EV Charging & Battery Swapping Infrastructure Value Chain map

Boundary

• InferencePublic distribution grid → station incoming line, box transformer / distribution → power input of charging equipment or the charging cabinets inside a swap station.

Interfaces and neighbours

• FactUpstream, grid companies are to "increase investment in supporting grid construction and reserve capacity for high-voltage, high-power charging".[P07 · （十六）]
• InferenceDownstream connects to C1–C2 on the charging branch and S1, S4 on the swapping branch.

Who pays

• InferenceThe station investor bears the electricity cost.
• FactFor grid companies' related distribution-network construction: "investment-efficiency constraints on grid companies' related distribution-network construction are relaxed, with full recovery through transmission and distribution tariffs".[P03 · 一、（三）]

Revenue and cost

• FactCentralised charging and swapping facilities on two-part tariffs are exempt from demand (capacity) charges before 2030.[P01 · 六、（二）][P03 · 一、（三）]
• FactPeak-valley time-of-use tariffs are to be implemented: P01 "guiding users to take part widely in smart orderly charging and vehicle-grid interaction"; P03 "encouraging users to charge in off-peak hours".[P01 · 六、（二）][P03 · 一、（四）]

Risks and constraints

• InferenceQueues for capacity upgrades and insufficient distribution-network capacity.
• GapThe demand-charge exemption reads "before 2030" (P01 original "2030年前", P03 original "到2030年前", both verified word for word); neither original says anything about afterwards, so there is no source.
• GapThe "outside the red line" investment boundary and the property demarcation point: the P07 original (byte-level search) contains none of "红线", "产权分界" or "接网"; no primary source says who pays grid-connection charges.

Verified source-record associations

No verified source-record association yet. This does not mean the stage has no companies, only that none is verified on this site.

Boundary

• InferenceCharging modules and control boards → complete DC chargers, AC chargers, shared-power charging stacks and liquid-cooled ultra-fast terminals.

Interfaces and neighbours

• FactCharging modules "are the core component of DC chargers, swap cabinets and storage-charging equipment".[C03 · p17]
• Fact盛弘股份 (300693) charger modules "cover 15kW, 20kW, 30kW, 40kW and 60kW power levels"; products include DC and AC chargers in integrated and split forms.[C02 · p13]
• InferenceThe same modules serve C1 and S1, and the direct buyer of modules is the charger maker. C03 shows neither a product supplied to both branches nor the module buyer; the top five customers are anonymous.

Who pays

• Fact"Rising investment by downstream charging and swapping operators drives demand for charging and swapping products".[C02 · p15]
• Fact绿能慧充 (600212) key-account channel is "mainly local state-owned urban, transport and industrial investment companies" (C04 is a newspaper disclosure page, secondary; the same sentence was verified on p19 of the CNINFO original summary 1223368158, pending source replacement).[C04 · B411 销售渠道]

Revenue and cost

• Fact通合科技 (300491) "charging power supplies for charging and swapping stations": 2024 revenue RMB 783 million (converted), 64.78% of revenue, gross margin 21.26% (down 6.32 points); one company's sample.[C03 · p25–26]
• Fact盛弘股份 (300693) "EV charging equipment": 2024 sales RMB 1.216 billion (converted), up 43.04%; the summary does not split chargers from modules.[C02 · p14]
• GapCharging-product gross margins of 盛弘, 绿能慧充, 万马 and 中恒 were not read; no source for market share.

Risks and constraints

• FactPower levels are moving up: 中恒电气 (002364) air-cooled DC chargers from 20kW to 1.44MW and liquid-cooled DC from 240kW to 720kW; 盛弘 lists a "1.2/1.44MW" stack product.[C06 · p4][C02 · p6]
• InferenceRisk of older products losing value; module gross margin declining (C03, one company).

Verified source-record associations

• 盛弘股份 (300693)ID 33Charging equipment maker: chargers and modules (one of its main products is "NEV charging and swapping equipment and services"; evidence from the summary only)[C02 · p12–13]
• 特锐德 (300001)ID 246Charging equipment manufacturing (secondary role; its charging business covers "charging equipment manufacturing, charging operation and value-added energy and data services")[C01 · p16]

Boundary

• InferenceSite + grid connection + equipment → an operable charging station; who owns the on-site equipment is this node's question.

Interfaces and neighbours

• InferenceUpstream: C1 and N1; downstream: C3.

Who pays

• InferenceThe investor's own capital expenditure.
• Fact"New expressway service areas should build charging infrastructure at the same time"; subject to safety, "public fast-charging and swapping facilities are to be added at suitable petrol (gas) stations".[P01 · 二、（一）][P01 · 三、（二）]

Revenue and cost

• GapPer-station investment, payback period and site-owner revenue share: all gaps, no primary data.

Risks and constraints

• InferenceLow utilisation leading to asset impairment.
• FactAn impairment case exists on the swapping side: 协鑫能科 (002015) "made asset impairments for some swapping and other projects with poor operating performance" (the original cites operating performance; the part read does not say "utilisation").[W03 · p24]
• GapA comparable impairment case on the charging side: gap.

Verified source-record associations

No verified source-record association yet. This does not mean the stage has no companies, only that none is verified on this site.

Boundary

• InferenceUser scans / app → charging starts → metering and settlement; two kinds: self-operating operators and aggregation platforms.

Interfaces and neighbours

• FactOperator sample: 特锐德 (300001) states it operated 709,000 public charging terminals at end-2024 (the annual report cites China EV Charging Infrastructure Promotion Alliance statistics) and charged over 13 billion kWh in 2024.[C01 · p18][C01 · p29]
• FactPlatform sample: NaaS announced in 2024 a strategic focus on a "platform - based interconnectivity charging business" (the original has spaces around the hyphen, reproduced as is), connecting users with "our charging station customers".[C07 · Item 3.D]
• GapWhether the connected stations belong to third-party operators and whether the platform holds no station assets: not stated in the part of C07 read; Item 4 not read.

Who pays

• InferenceVehicle owners (electricity plus service fee); no service-fee pricing document was read.

Revenue and cost

• Fact特锐德 (300001) "EV charging": 2024 revenue RMB 4.890 billion (converted), 31.80% of revenue, gross margin 33.24%; the line's cost item is "product and service cost", covering equipment and services, so it is not pure operating revenue.[C01 · p31–32]
• Fact特锐德 changed charging-operation revenue recognition "from gross to net" with retrospective adjustment.[C01 · p8]
• InferenceThat revenue therefore cannot be compared directly with peers reporting on a gross basis.
• FactNaaS net loss of RMB 914.4 million in 2024.[C07 · Item 3.D 风险因素]
• GapService-fee price, electricity share of operating cost and average station utilisation: all gaps; C07 Items 4/5 not read.

Risks and constraints

• FactIn late August 2023 "three major charging station operators in China simultaneously terminated their collaboration with us", disconnecting "a meaningful percentage" of the stations and ports on the NaaS network.[C07 · Item 3.D]
• InferenceA platform's network scale is decided unilaterally by operators (author's summary).

Verified source-record associations

• 特锐德 (300001)ID 246Charging operator (primary role): 709,000 public charging terminals operated (citing alliance statistics); charging revenue on a net basis. 特来电 is a 77.71%-held subsidiary; treating it as the operating entity is an inference.[C01 · p5、p18、p31]

Boundary

• InferenceEquipment inspection, fault handling, software upgrades and contracted operation.

Interfaces and neighbours

• Fact万马股份 (002276) "has expanded into contracted operation and contracted O&M services".[C05 · 全文印刷 p10（PDF 第 11 页）]
• Fact特锐德 (300001) charging business covers "charging equipment manufacturing, charging operation and value-added energy and data services".[C01 · p16]
• Fact绿能慧充 (600212) business includes "charging station construction, operation and maintenance" (C04 is a newspaper disclosure page, secondary; the same sentence was verified on p5 of the CNINFO original summary 1223368158, pending source replacement).[C04 · B411 业务概况]

Who pays

• InferenceOperators or station investors.

Revenue and cost

• GapNo separate amount read; 万马's "trade, services and other" line cannot be equated with charging-service revenue.

Risks and constraints

• InferenceO&M quality affects availability.
• GapForm of O&M contracts and charging method: gap.

Verified source-record associations

• 特锐德 (300001)ID 246Value-added energy and data services (secondary role; primary/secondary ordering is the author's, pending review)[C01 · p16]

Boundary

• InferenceSwap robots, station body, in-station charging cabinets and battery boxes → whole-station delivery.

Interfaces and neighbours

• FactUpstream: charging modules are the core component of "swap cabinets" (equating swap cabinets with in-station charging cabinets is an inference).[C03 · p17]
• Fact山东威达 (002026) sub-subsidiary 昆山斯沃普 "has supplied third- and fourth-generation swap station equipment to NIO".[W05 · p12]
• Fact博众精工 (688097) smart charging / swapping station products cover passenger-vehicle and commercial-vehicle stations, battery boxes and chargers; customers include 宁德时代, SAIC, PetroChina, Sinopec and NIO.[W06 · p15]
• InferenceDownstream: S4 operators and vehicle makers.

Who pays

• FactOne of two revenue bases: 山东威达 works on a processing basis — "downstream customers supply the main materials and the company provides processing and charges a processing fee".[W05 · p14]
• InferenceThe payer is the operator or the vehicle maker.

Revenue and cost

• GapThe other basis (equipment sales): the 2025 HKEX application proof cited as W04 returned HTTP 404 on 2026-10-21 and 2026-10-23, so "equipment sales RMB 268 million in 2024, RMB 731 million in 2022" cannot be verified against the original and is downgraded to a gap. Cross-check: the 2026-07-19 application proof (hkexnews, SHA-256 a117a173…, printed p2) shows equipment sales of RMB 267,999 thousand (28.9%) in 2024 and RMB 519,174 thousand in 2023, with no 2022 figure; to be restored as a fact once the source table is switched to that version and re-reviewed.
• GapSwap-equipment revenue of 博众 and 瀚川 not read; 山东威达's "new energy products" RMB 809 million (p17) corresponds to the "energy-storage battery pack business and NEV swap-station business" described on p13, so it includes battery packs and cannot be taken as swap-equipment revenue.[W05 · p13、p17]

Risks and constraints

• InferenceCustomer concentration: equipment revenue follows a few operators' station plans; the part of W05 read names only NIO as a swap-equipment customer.
• Fact瀚川智能 (688022) made a "strategic contraction of the battery equipment and charging / swapping equipment businesses".[W07 · 2.3 节（PDF 第 6 页）]

Verified source-record associations

No verified source-record association yet. This does not mean the stage has no companies, only that none is verified on this site.

Boundary

• InferencePack dimensions, locking and interfaces → the set of vehicle models interchangeable at one station.

Interfaces and neighbours

• FactNational standard GB/T 40032-2021 Safety requirements of battery swap for electric vehicles, issued by SAMR and the Standardization Administration of China, published 2021-04-30 and in force from 2021-11-01 (information page only; text not read; drafters are not on that page and are not stated here).[P05 · 标准信息页]
• FactCompany scheme: 宁德时代 (300750) "launched the 20# and 25# standard swap blocks", formed the "Choco-Swap alliance" with a hundred partners and co-launched ten swap-compatible models.[W02b · 印刷 p32（PDF 第 35 页）]
• FactPolicy goal: "promote unified swapping standards in the main application areas".[P07 · 三、（十二）]

Who pays

• InferenceVehicle makers bear the cost of model adaptation.

Revenue and cost

• GapNo separate revenue disclosure for standards or the alliance; the 宁德时代 revenue table read does not list swapping separately, and "other business" cannot be taken as swapping revenue.[W02 · p19]

Risks and constraints

• InferenceFragmented standards are the core risk of the swapping branch: a station serves only compatible models; P05 is a safety requirement and does not unify pack dimensions.
• GapInteroperability between swapping standards and the number of compatible models: gap. Drafters: the std.samr.gov.cn standard detail page lists 上海蔚来汽车有限公司 first among "main drafting organisations" (fetched 2026-10-23, SHA-256 e302a36a…); that page is not in this mapping's source table and awaits review before being added.

Verified source-record associations

• 宁德时代 (300750)ID 242Battery-standard scheme provider (candidate): 20# / 25# standard swap blocks and the Choco-Swap alliance. Its S4 operator role is not listed until evidence of the actual operating entity is obtained; in S3 it appears only as an initial investor in 蔚能.[W02b · 印刷 p32（PDF 第 35 页）]

Boundary

• InferenceBuy batteries → hold battery assets → lease / subscribe them to users.

Interfaces and neighbours

• FactNIO BaaS: "For each user under the BaaS model, we sell a battery to Wuhan Weineng Battery Asset Co., Ltd., … and the user subscribes for the usage of the battery from the Battery Asset Company."[W01 · Item 3.D]
• FactAs of the annual report date NIO owned about 19.4% of 蔚能; 宁德时代 was one of its initial investors.[W01 · Item 3.D]

Who pays

• FactBaaS users subscribe for battery usage from the Battery Asset Company and "pay a monthly subscription fee for the battery" (the amount was not read).[W01 · Item 3.D]

Revenue and cost

• InferenceThe asset holder collects subscription fees; its costs are battery depreciation and financing.
• FactNot to be conflated: NIO's 2024 related-party "sales of goods" to 蔚能 and its subsidiary of RMB 9.848 billion (converted) are a vehicle maker's sales to the asset holder, not swapping-service revenue; R118 does not say the goods are batteries.[W01 · R118]

Risks and constraints

• InferenceBattery residual value and technology turnover.
• FactAs of 2024-12-31 the guarantee liability NIO provided to 蔚能 was "immaterial".[W01 · Item 3.D]
• Gap蔚能's own financial statements not obtained; subscription price and battery depreciation life: gaps.

Verified source-record associations

No verified source-record association yet. This does not mean the stage has no companies, only that none is verified on this site.

Boundary

• InferenceVehicle enters → depleted battery swapped for a charged one → batteries recharged in station.

Interfaces and neighbours

• FactVehicle maker: NIO's power solutions include "battery swapping, which we refer to as Power Swap", developed independently and with partners; who owns and runs each station is not stated in the part read.[W01 · Item 3.D]
• GapIndependent third party (奥动) station count: the 2025 application proof cited as W04 returned HTTP 404 on both fetches, so "267 own stations, 62 operating-service and 192 platform-service third-party stations at 2025-06-30" cannot be verified and is downgraded to a gap. Cross-check: the 2026-07-19 application proof (SHA-256 a117a173…, printed p37) reports 321, 291, 236 and 214 own swap stations at end-2023, end-2024, end-2025 and 2026-04-30, a different date, so it does not replace the cited quote.
• FactBattery-maker scheme: 宁德时代 (300750) launched the Choco-Swap and Qiji swapping solutions and set up 时代电服（江苏）科技有限公司.[W02 · p15][W02 · p22]
• GapThe actual swap-operating entity of 宁德时代 is unverified; it is not listed as a member of this node.

Who pays

• InferenceVehicle owners or fleets pay per swap or per kilometre; the pricing method was not read.

Revenue and cost

• Gap奥动 "swapping services at own stations" RMB 547 million in 2024, 59.0% of revenue: the 2025 application proof cited as W04 returned HTTP 404 on both fetches, so this cannot be verified and is downgraded to a gap. Cross-check: the 2026-07-19 application proof (SHA-256 a117a173…, printed p2) shows RMB 546,761 thousand, 59.0% of total revenue, for 2024, the same figures; to be restored as a fact once the source table is switched to that version and re-reviewed.
• FactNIO "Provision of power solution": RMB 2.101 billion in 2024 (converted), separate from vehicle sales; its composition was not read and by name it may include charging services, so it is not pure swapping revenue.[W01 · R93]
• Gap协鑫能科's product revenue table has no swapping line; swapping revenue not read.[W03 · p17]

Risks and constraints

• Fact协鑫能科 (002015) "made asset impairments for some swapping and other projects with poor operating performance"; the impairment is not broken out for swapping.[W03 · p16][W03 · p24]
• Gap奥动 total revenue falling from RMB 1.155 billion (2023) to RMB 926 million (2024), mainly equipment sales by the author's calculation: the 2025 application proof cited as W04 returned HTTP 404 on both fetches, so this cannot be verified and is downgraded to a gap. Cross-check: the 2026-07-19 application proof (SHA-256 a117a173…, printed p2) shows total revenue of RMB 1,155,368 thousand (2023) and 926,292 thousand (2024) and equipment sales of 519,174→267,999 thousand, the same figures; to be restored after the source is switched and re-reviewed.
• GapNational swap-station count: the P06 / P06b originals (byte-level search) contain no "换电"; swaps per station and electricity cost: gaps. The 2021 pilot notice text was not obtained; pilot cities and targets appear only in re-posts (P04, secondary).

Verified source-record associations

• 协鑫能科 (002015)ID 69Swap operation (structural inference: the part read only has swapping-project impairments and R&D wording; swapping revenue and station count not read)[W03 · p16、p24]

Boundary

• InferenceSwap equipment maintenance, battery testing, second-life use and recycling.

Interfaces and neighbours

• Gap奥动 "swap operating services" (day-to-day operation, maintenance, customer support and / or platform services for swap stations) RMB 85 million in 2024: the 2025 application proof cited as W04 returned HTTP 404 on both fetches, so this cannot be verified and is downgraded to a gap. Cross-check: the 2026-07-19 application proof (SHA-256 a117a173…, printed p2) shows RMB 84,728 thousand (9.1%) for 2024 with the same service description; to be restored after the source is switched and re-reviewed.

Who pays

• Gap奥动's statement that "customers buying our swap stations and operating services are mainly local urban investment companies, energy investment companies, transport enterprises and grid companies": the 2025 application proof cited as W04 returned HTTP 404 on both fetches, so this cannot be verified and is downgraded to a gap. Cross-check: the 2026-07-19 application proof (SHA-256 a117a173…, printed p4) carries the same sentence; to be restored after the source is switched and re-reviewed.
• InferenceTreating these customers as the payers for operating services is an inference; the customer mix itself is unverified because the cited application proof returned 404, see the gap above.

Revenue and cost

• GapCost structure and contract form of operating services not read.

Risks and constraints

• InferenceBattery health management determines asset value.
• GapPrimary sources on second-life use and recycling not checked (interface to the battery / storage chain).

Verified source-record associations

No verified source-record association yet. This does not mean the stage has no companies, only that none is verified on this site.

Boundary

• InferenceSmart orderly charging and bidirectional charging-discharging; aggregated participation in power-market pilots and demand response, spanning N1, C3 and S4.

Interfaces and neighbours

• Fact"Vehicle-grid interaction mainly includes smart orderly charging and bidirectional charging-discharging, and can take part in peak shaving, virtual power plants and aggregated trading."[P02 · 引言（第二段）]
• Fact"Bidirectional facilities, storage-charging / PV-storage-charging stations and swap stations are encouraged to join power-market pilots through resource aggregation".[P02 · 二、（三）]
• Fact"Promote pilots of connected vehicles, vehicle-grid interaction, source-grid-load-storage integration and PV-storage-charging-swapping stations".[P01 · 五、（一）]

Who pays

• InferenceThe peak-valley price spread accrues to vehicle owners or operators.
• GapNo source on how aggregated-trading income is shared.

Revenue and cost

• Fact"Aim to build more than 5 demonstration cities and more than 50 bidirectional charging-discharging demonstration projects by end-2025."[P02 · 三、（三）]
• GapActual returns of demonstration projects and aggregator settlement: gaps.

Risks and constraints

• InferenceIf settleable aggregated income emerges, charging operators and swap stations gain a grid-service revenue line; pending settlement data from pilots.
• GapThe P01 and P02 originals (byte-level search) do not use "V2G"; they say "bidirectional charging-discharging" and "vehicle-grid interaction". The P03 original, 1(4), has the sentence "encourage key technology research on electric-vehicle–grid bidirectional interaction (V2G) and PV-storage-charging coordinated control". Research note: company samples for this chain are retrieved by the power grid equipment, energy services and energy storage industries; an industry match does not prove charging or swapping business and must be checked against primary sources each round. Verified associations on this site are only source records 246, 33, 242 and 69.

Verified source-record associations

No verified source-record association yet. This does not mean the stage has no companies, only that none is verified on this site.