Storage carbon neutralization dynamic optimization method based on multi-modal data fusion

Abstract

The invention discloses a storage carbon neutralization dynamic optimization method based on multi-modal data fusion, and solves the problem that carbon neutralization cannot be realized through a single carbon emission reduction measure due to the fact that storage estimated carbon emission is in dynamic change in the prior art. Formulating a carbon emission reduction plan and a carbon emission plan, and calculating the planned carbon emission; warehousing operation data and operation data are obtained, and estimated carbon emission is calculated; if the estimated carbon emission exceeds the planned carbon emission, necessary carbon emission is calculated, if the necessary carbon emission does not exceed the planned carbon emission, transportation and storage are carried out according to an optimization scheme, and otherwise, an emission increasing and reducing suggestion is generated according to an emission increasing and reducing optimization strategy. According to the method, the pre-estimated carbon emission in the warehouse-in and warehouse-out stage is pre-estimated, the necessary carbon emission is calculated to decide whether a carbon emission reduction measure needs to be added or not, and meanwhile, the operation path of transportation equipment and the storage position of goods are optimized, so that the storage carbon emission is effectively reduced.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of carbon neutrality, and in particular relates to a warehouse carbon neutrality dynamic optimization method based on multi-modal data fusion. BACKGROUND

[0002] Carbon neutrality generally refers to the total amount of carbon dioxide or greenhouse gas emissions directly or indirectly generated by a country, enterprise, product, activity or individual within a certain period of time, which is offset by afforestation, energy saving and emission reduction, etc. to offset the carbon dioxide or greenhouse gas emissions generated by itself, achieve positive and negative offset, and achieve relative "zero emission".

[0003] In the current electric power material warehouse, in order to realize the carbon flow closed loop of zero carbon warehouse park and realize carbon neutrality, it is necessary to detect and analyze the carbon emission data in the park, and to count the carbon emission reduction measures in the park. The energy resource consumption data of electric power material warehouse, warehouse transfer inspection operation and daily operation is monitored. The current mainstream emission reduction measures in the park include purchasing green plants, and estimating the emission reduction amount by converting carbon sink through green plants, so that the estimated carbon emission reduction amount and the estimated carbon emission amount are neutralized. However, in actual application, the carbon emission amount is affected by many factors, such as the running state of the equipment, the amount of warehouse materials and the type of warehouse, etc. The estimated carbon emission amount is in dynamic change, and carbon neutrality cannot be achieved by a single carbon emission reduction measure. SUMMARY

[0004] The purpose of the present application is to solve the problem that the estimated carbon emission amount in the warehouse is in dynamic change in the prior art, which leads to the fact that carbon neutrality cannot be achieved by a single carbon emission reduction measure. A warehouse carbon neutrality dynamic optimization method based on multi-modal data fusion is provided, which estimates the estimated carbon emission amount in the warehouse and calculates the necessary carbon emission amount to decide whether to increase the carbon emission reduction measure. At the same time, the running path of the transportation equipment and the storage position of the goods are optimized to effectively reduce the warehouse carbon emission amount.

[0005] In order to achieve the above-mentioned purpose, the present application adopts the following technical scheme: A warehouse carbon neutrality dynamic optimization method based on multi-modal data fusion, characterized in that it comprises the following steps: S1: According to the historical data, the planning period is divided, the phased carbon emission reduction plan and the phased carbon emission plan are formulated, and the planned carbon emission amount is calculated; S2: Obtain warehouse operation data and operation data, and calculate the estimated carbon emission amount in the current planning period; S3: In the current planning period, if the estimated carbon emissions exceed the planned carbon emissions, calculate the necessary carbon emissions, if the necessary carbon emissions do not exceed the planned carbon emissions, transport and store according to the optimization scheme, otherwise generate emission reduction and increase suggestions according to the emission reduction and increase optimization strategy.

[0006] The method of the application realizes carbon neutralization periodically by setting pre-planning, and pre-estimates the estimated carbon emissions of the warehouse-in and warehouse-out stage and calculates the necessary carbon emissions to decide whether to increase carbon emission reduction measures, while optimizing the running path of the transportation equipment and the storage position of the goods to reduce unnecessary carbon emissions.

[0007] As a preferred, the calculating necessary carbon emissions comprises: obtaining goods information and goods storage information; combining a plurality of devices to develop a plurality of simulation schemes respectively, simulating the transportation state and operation data of each transportation device, and calculating the transportation carbon emissions of the transportation process in each simulation scheme; selecting the simulation scheme with the minimum transportation carbon emissions as the optimization scheme, and the sum of the transportation carbon emissions and the operation carbon emissions corresponding to the optimization scheme is the necessary carbon emissions.

[0008] As a preferred, the emission reduction and increase optimization strategy comprises: obtaining emission reduction measures and calculating estimated carbon emission reduction, taking the difference between the necessary carbon emissions and the estimated carbon emission reduction as the new emission reduction amount; obtaining the current emission reduction measures and the preset carbon emission reduction cost corresponding to each emission reduction measure, generating a plurality of emission reduction measure combinations and calculating the corresponding estimated emission reduction cost; outputting a plurality of emission reduction measure combinations and estimated emission reduction cost.

[0009] As a preferred, it comprises transportation equipment abnormality detection: obtaining abnormal task conditions, when the on-off time of the transportation equipment is greater than a preset first time threshold, performing task statistics and abnormal task condition judgment, if the abnormal task condition is detected, an optimization task flow signal is generated.

[0010] As a preferred, it comprises warehouse operation abnormality detection: obtaining abnormal power consumption, if abnormal power consumption is detected, an optimization operation power signal is generated, the abnormal power consumption comprises being in an abnormal lighting period and the continuous lighting time exceeding a preset lighting time threshold, or being in an abnormal power consumption power period and the abnormal power duration exceeding a preset power duration threshold.

[0011] As a preferred, the running state prediction model is trained, the running path and running condition of the transportation equipment are taken as the output, and the storage operation data is outputted.

[0012] As a preferred, the emission reduction and increase optimization strategy is configured with an efficacy correction sub-strategy, the efficacy correction sub-strategy corrects the corresponding optimization coefficient according to the use time of the emission reduction measure.

[0013] Preferably, the abnormal task conditions include standby time exceeding a second time threshold or idle time exceeding a third time threshold, and the idle time accounting for a preset percentage of the total task time.

[0014] Preferably, the operation status prediction model is configured with a model correction strategy, which compares the actual transportation data with the operation data and corrects the operation status prediction model when the difference between the actual transportation data and the operation data is greater than a preset error threshold.

[0015] Preferably, the operational carbon emissions are obtained by converting energy consumption data from historical data on building electrical equipment consumption and operational material consumption records into carbon emissions.

[0016] Therefore, the present invention has the following beneficial effects: by setting pre-planning steps, carbon emissions and carbon reduction are planned in advance, so as to achieve carbon neutrality periodically. It also considers the carbon emissions during the inbound and outbound stages and the operation stage simultaneously, estimates the estimated carbon emissions during the inbound and outbound stages and calculates the necessary carbon emissions to decide whether to increase carbon reduction measures. At the same time, by optimizing the operation path of transportation equipment and the storage location of goods, unnecessary carbon emissions are reduced, thus achieving carbon neutrality in warehousing. Attached Figure Description

[0017] Figure 1 This is a flowchart illustrating the overall steps of the dynamic optimization method for carbon neutrality in storage based on multimodal data fusion in this invention.

[0018] Figure 2 This is a schematic diagram of the optimized path for carbon neutralization in this invention. Detailed Implementation

[0019] The present invention will be further described in detail below with reference to the accompanying drawings and specific embodiments: Example 1: This embodiment provides a dynamic optimization method for warehouse carbon neutrality based on multimodal data fusion, such as... Figure 1 As shown, the operation process is as follows: Step 1, divide the planning period according to historical data, formulate phased carbon emission reduction plans and phased carbon emission plans, and calculate the planned carbon emissions; Step 2, obtain warehousing operation data and operational data, and calculate the estimated carbon emissions in the current planning period; Step 3, in the current planning period, if the estimated carbon emissions exceed the planned carbon emissions, calculate the necessary carbon emissions; if the necessary carbon emissions do not exceed the planned carbon emissions, carry out transportation and warehousing according to the optimized plan; otherwise, generate emission reduction suggestions based on the emission reduction optimization strategy.

[0020] This embodiment provides a dynamic optimization method for carbon neutrality in warehousing based on multimodal data fusion. By pre-planning carbon emissions and reductions, it achieves periodic carbon neutrality. It simultaneously considers carbon emissions during the inbound / outbound and operational phases, pre-estimating the predicted carbon emissions during inbound / outbound and calculating the necessary carbon emissions to decide whether additional carbon reduction measures are needed. Furthermore, it optimizes the operating routes of transportation equipment and the storage locations of goods to reduce unnecessary carbon emissions. The following specific examples and application scenarios further illustrate the technical solution and effects of this invention. These examples are illustrative of the invention, and the invention is not limited to these examples.

[0021] Specifically, this manifests as follows: Step 1: Divide the planning period based on historical data, formulate phased carbon reduction plans and phased carbon emission plans, and calculate the planned carbon emissions.

[0022] The phased carbon emission reduction plan includes planned carbon emission reductions, and the phased carbon emission plan includes planned carbon emissions. The sum of the planned carbon emission reductions within one or more consecutive planning periods is equal to the sum of the planned carbon emissions.

[0023] Step 2: Obtain warehouse operation data and business data, and calculate the estimated carbon emissions for the current planning cycle.

[0024] Step 3: If the estimated carbon emissions exceed the planned carbon emissions within the current planning period, calculate the necessary carbon emissions. If the necessary carbon emissions do not exceed the planned carbon emissions, proceed with transportation and storage according to the optimized plan. Otherwise, generate emission increase / decrease recommendations based on the emission increase / decrease optimization strategy.

[0025] This involves obtaining the estimated carbon emissions and estimated carbon reductions for the current planning period. When the estimated carbon emissions exceed the planned carbon emissions, the necessary carbon emissions are calculated based on the necessary carbon emission calculation strategy. If the necessary carbon emissions do not exceed the planned carbon emissions, transportation and storage are carried out according to the optimized plan. If the necessary carbon emissions still exceed the planned carbon emissions, emission increase and decrease suggestions are generated based on the emission increase and decrease optimization strategy.

[0026] The calculation of necessary carbon emissions based on the necessary carbon emission calculation strategy includes: (1) Obtain cargo information and cargo storage information, including cargo entry and exit time, cargo weight and cargo storage location.

[0027] (2) Combine multiple devices to formulate multiple simulation schemes. Each simulation scheme is configured with different transportation equipment and transportation routes to simulate the transportation process of each transportation device.

[0028] Specifically, it includes: Simulate the operating status of various transportation devices, including standby, use, and charging.

[0029] Simulate the operational data of various transportation devices, including transportation trajectory, transportation time, and transportation power, and calculate the transportation carbon emissions during the simulated transportation process.

[0030] (3) Compare the transportation carbon emissions under each simulation scheme, and take the simulation scheme with the smallest transportation carbon emissions as the optimization scheme for storage carbon neutrality optimization. The sum of the transportation carbon emissions and the operation carbon emissions corresponding to the optimization scheme is taken as the necessary carbon emissions.

[0031] Among them, the operating carbon emissions are obtained by converting energy into carbon through monitoring the energy consumption of building electrical equipment and the consumption records of operating materials in historical data.

[0032] Furthermore, emission reduction and optimization strategies include: The emission reduction optimization strategy includes calculating the difference between the necessary carbon emissions and the estimated carbon emission reduction as the new emission reduction, obtaining the current emission reduction measures and the preset carbon emission reduction costs corresponding to each emission reduction measure, generating multiple emission reduction measure combinations and calculating the corresponding estimated emission reduction costs. The emission reduction measure combination includes the selected emission reduction measures and their corresponding proportions, and outputs the multiple emission reduction measure combinations and estimated emission reduction costs respectively.

[0033] The emission reduction measures include increasing solar power and increasing green vegetation.

[0034] Meanwhile, the emission reduction optimization strategy is also equipped with an effectiveness correction sub-strategy. The effectiveness correction sub-strategy includes adjusting the corresponding optimization coefficient according to the usage time of the emission reduction measures. When the time is summer, the optimization coefficients for increasing photovoltaic power and increasing greening will be increased accordingly.

[0035] This embodiment also includes: The pre-trained operation status prediction model takes the equipment's operating path and transportation conditions as input and outputs warehouse operation data.

[0036] The training data for the operational status prediction model includes actual transportation routes, actual output power, actual working hours, actual load, and operational status.

[0037] In practical applications, the actual transportation path of the transportation equipment can be collected by the equipment operation trajectory collector, the actual output power of the transportation equipment can be obtained by the equipment operation power collector, the actual working time of the transportation equipment can be obtained by the equipment operation time collector, the actual load of the transportation equipment can be collected by the load sensor, and the operating status of the transportation equipment can be obtained by the equipment status detector.

[0038] Furthermore, the operation status prediction model is configured with a model correction strategy. The model correction strategy obtains the actual transportation route, actual output power, actual working time, actual load and operation status, and compares them with the operation data, including transportation trajectory, transportation time and transportation power. If the difference is greater than a preset error threshold, the operation status prediction model is corrected.

[0039] In this embodiment, the dynamic optimization method for carbon neutrality in storage also includes equipment anomaly detection: Equipment anomaly detection includes acquiring abnormal task conditions. When the time from device startup to shutdown exceeds a preset first time threshold, task statistics are performed, and abnormal task conditions are judged. If abnormal task conditions are detected, the task is identified as abnormal and an optimized task process signal is generated.

[0040] Abnormal task conditions include standby time exceeding the second time threshold or idle time exceeding the third time threshold, and the idle time accounting for a preset percentage of the total task time.

[0041] In this embodiment, the dynamic optimization method for carbon neutrality in warehousing also includes warehouse operation anomaly detection: Warehouse operation anomaly detection includes acquiring abnormal power consumption. If abnormal power consumption is detected, it is identified as an operational anomaly, and an optimized power consumption signal is generated.

[0042] Abnormal power consumption includes periods of abnormal lighting with continuous lighting duration exceeding a preset lighting duration threshold, and periods of abnormal power consumption with abnormal power duration exceeding a preset power duration threshold.

[0043] As can be seen from the above, the storage carbon neutrality dynamic optimization method based on multimodal data fusion provided in this embodiment has the following optimization path: Figure 2 As shown, the optimization path for carbon neutrality in warehousing includes two paths: one is to increase carbon emission reductions, and the other is to reduce carbon emissions. These approaches address different aspects to achieve carbon neutrality optimization in warehousing.

[0044] Specifically: (1) Increase carbon emission reduction.

[0045] Increasing carbon emission reductions involves setting carbon neutrality targets, configuring the proportion of green electricity trading and the proportion of carbon sinks in park green plants when the targets are achieved, and then outputting green electricity trading schemes and park green plant carbon sink schemes (including targets and costs).

[0046] (2) Reduce carbon emissions.

[0047] Reducing carbon emissions includes: Data monitoring of warehousing equipment reveals abnormal operating data, indicating equipment malfunction and requiring optimization. Optimization includes adjusting standby time, equipment usage, charging time, and routing to reduce carbon emissions.

[0048] If warehouse operational anomalies are detected, the warehouse needs to be optimized. Optimization includes improvements to lighting, electrical outlets, air conditioning, and power systems to reduce carbon emissions.

[0049] Example 2: This embodiment provides a dynamic optimization method for carbon neutrality in warehouses based on multimodal data fusion, and further optimizes the solution in Embodiment 1 by applying it to a specific application scenario.

[0050] A dynamic optimization method for carbon neutrality in warehouses based on multimodal data fusion includes: (a) Pre-planning steps.

[0051] The preliminary planning steps divide planning periods based on historical data and formulate phased carbon emission reduction plans and phased carbon emission plans. The phased carbon emission reduction plan includes the planned carbon emission reduction amount, and the phased carbon emission plan includes the planned carbon emission amount. The sum of the planned carbon emission reduction amounts in one or more consecutive planning periods is equal to the sum of the planned carbon emission amounts.

[0052] Specifically: Multiple planning periods can be set, such as using an annual planning unit, dividing the year into multiple planning periods by quarters. The expected carbon emission reduction and the estimated carbon emission in each planning period can be different. At the annual settlement, the total estimated carbon emission and the total estimated carbon emission reduction should be equal. Calculating the total over a certain time span can avoid frequent updates and optimizations of the plan due to equipment failures or data anomalies in a short period.

[0053] New phased carbon emission plans can be formulated based on historical carbon emission data. For example, if the carbon emission data shows that building electricity consumption occurs in summer and winter, or that frequent logistics and warehousing scheduling in certain months significantly increases carbon emissions, then a larger estimated amount of carbon emissions can be allocated to the corresponding planning period. Specifically, this can be calculated in the form of a percentage.

[0054] (ii) Carbon emission reduction estimation steps.

[0055] The carbon emission reduction estimation process includes obtaining emission reduction measures and calculating the estimated carbon emission reduction.

[0056] (iii) Carbon emission estimation steps.

[0057] The carbon emissions estimation process includes acquiring warehouse operation data and operational data to calculate the estimated carbon emissions for the current planning period.

[0058] (iv) Optimization suggestions and steps.

[0059] The optimization suggestion steps include obtaining the estimated carbon emissions and estimated carbon emission reductions for the current planning period. When the estimated carbon emissions exceed the planned carbon emissions, the necessary carbon emissions are calculated according to the necessary carbon emission calculation strategy. If the necessary carbon emissions do not exceed the planned carbon emissions, transportation and storage are carried out according to the optimization plan. If the necessary carbon emissions still exceed the planned carbon emissions, emission increase and decrease suggestions are generated according to the emission increase and decrease optimization strategy.

[0060] Calculating the necessary carbon emissions includes: Obtain information on goods, including their entry and exit times, weight, and storage location. Multiple simulation schemes are developed by combining multiple devices. Each simulation scheme is configured with different transportation equipment and routes, and simulates the transportation process of each piece of equipment. The operating states of various transportation devices are simulated, including standby, use, and charging states. The operational data of various transportation devices, including transportation trajectory, transportation time, and transportation power, are simulated to calculate the carbon emissions during the simulated transportation process. The carbon emissions from transportation are compared under each simulation scheme, and the simulation scheme with the lowest carbon emissions from transportation is taken as the optimized scheme. The sum of the carbon emissions from transportation and the carbon emissions from operation corresponding to the optimized scheme is taken as the necessary carbon emissions.

[0061] Among them, the operating carbon emissions are obtained by converting energy into carbon through monitoring the energy consumption of building electrical equipment and the consumption records of operating materials in historical data.

[0062] In this embodiment, carbon emissions in the warehousing park mainly come from two stages: the inbound and outbound transportation of goods and the storage and operation of goods in the warehouse.

[0063] The overall goal of the warehouse carbon neutrality optimization scheme is to collect detailed data on each batch of goods (task time, task equipment, goods type, goods entry / exit time, goods weight, goods batch, goods location, etc.) to ultimately optimize the entire operational process. This includes simplifying processes, improving inventory management, and reducing unnecessary goods movement (reducing goods transportation).

[0064] A pre-trained operation status prediction model is used, which takes the equipment's operating path and transportation conditions as input and outputs the operation data.

[0065] This embodiment also includes device anomaly detection and acquisition of abnormal task conditions. When the time from device power-on to device termination exceeds a preset first time threshold, task statistics are performed and abnormal task conditions are judged. If abnormal task conditions are detected, the task is identified as abnormal and an optimized task flow signal is generated. The abnormal task conditions include standby time exceeding a second time threshold or idle time exceeding a third time threshold and the idle time accounting for a preset percentage of the total task time.

[0066] When retrieving abnormal tasks, the specific steps include: (1) Set the data collection indicators.

[0067] In this embodiment, the collected indicators include task start time, task end time, real-time voltage, real-time current, real-time power, and total power consumption.

[0068] The indicators that need to be collected on-site include: Standby power (range), full load power (range), and no-load power (range).

[0069] (2) Conditions for confirming abnormal tasks.

[0070] (2.1) For each task whose duration from machine startup time to end time is greater than 5 minutes, task statistics will be performed; (2.2) Real-time power (standby phase): The standby time is greater than 10 minutes for each task; (2.3) Real-time power (no-load phase): The no-load phase lasts longer than 10 minutes for each task and accounts for more than 50% of the total task duration.

[0071] In this embodiment, the warehouse carbon neutrality optimization method includes reducing carbon emissions from automated equipment and reducing carbon emissions from manually operated equipment.

[0072] Specifically: Reducing carbon emissions from automated equipment includes: statistically analyzing the duration of different operating phases of all automated equipment; analyzing the data through a platform to determine the normal (effective emissions and ineffective emissions) of the automated equipment; and reducing ineffective emissions from the equipment through technological means.

[0073] Carbon emission reduction for manually operated equipment includes: tracking the duration of different transportation stages of manually operated equipment; analyzing the data through a platform to determine the normal (effective and ineffective emissions) of automated equipment; and reducing ineffective emissions of equipment through behavioral norms.

[0074] In this embodiment, the warehouse carbon neutrality optimization scheme includes carbon emission monitoring, carbon emission reduction monitoring, and carbon neutrality construction.

[0075] Furthermore, carbon emission monitoring includes: Inbound / outbound stage: Each batch of goods in the warehouse management system is linked to the corresponding handling equipment, refining the carbon footprint acquisition to the specific batch. Specifically: The warehouse management system (WMS) is used to record detailed batch information and inbound / outbound details; an equipment operation data acquisition system is used to collect data on equipment operating distance, trajectory, power consumption, and runtime, including gantry cranes, cranes, forklifts, stacker cranes, and AGVs.

[0076] During the warehousing phase: Carbon emissions from energy consumption and material consumption during this phase are statistically analyzed. This data is then refined down to the batch level through the integrated WMS (Building Management System). Specifically, a smart building control system is used to monitor energy consumption at the equipment level, including air conditioners, sockets, lighting, and elevators. Carbon conversion coefficients are based on national standards. An office supplies ledger system is used to record the carbon emissions of office consumables, including packaging, office supplies, and paper. Carbon conversion coefficients are based on national standards.

[0077] Furthermore, carbon emission reduction monitoring includes: real-time monitoring of the overall carbon emission reduction situation in the park and various carbon emission reduction data.

[0078] Specifically, it includes: Photovoltaic construction: A photovoltaic power generation data acquisition system is used to collect photovoltaic power generation data. The acquisition methods include data collection through smart meters, data collection by connecting to the photovoltaic power generation system, and photovoltaic data reporting.

[0079] Green electricity consumption: The green electricity data accounting system of the warehouse park is used to collect green electricity data in the circular area. The collection methods include connecting the park's main electricity meter to the collector to collect the total electricity consumption and manually filling in the total electricity consumption data.

[0080] Purchasing Green Certificates: Green certificate data collection, including manual entry and data conversion according to national standards.

[0081] Green Plant Conversion: This function aggregates carbon emission data for green plants, using methods including manual data entry and data conversion based on national standards.

[0082] Furthermore, carbon neutrality construction includes: Carbon neutrality status analysis and forecast for the industrial park: This section displays and analyzes current carbon emissions and total carbon reduction data for the park, as well as information on materials and detailed equipment operation data. This can be presented in the form of an analysis report or in real-time in the cockpit of the transportation equipment.

[0083] Carbon Neutrality Construction Plan: This plan analyzes the current emission reduction and carbon emission reduction reasons in the park, and displays material information, detailed equipment operation data, and the overall carbon neutrality optimization plan. The carbon neutrality optimization plan is presented in report form, including plans to increase or decrease carbon emission reduction and plans to further reduce carbon emissions.

[0084] Carbon neutrality pathways include green electricity purchases, photovoltaic expansion, green planting, and green certificate trading.

[0085] The embodiments described above are merely preferred embodiments of the present invention and are not intended to limit the present invention in any way. Other variations and modifications are possible without departing from the technical solutions described in the claims.

Claims

1. A dynamic optimization method for carbon neutrality in warehouses based on multimodal data fusion, characterized in that, include: S1: Divide the planning cycle according to historical data, formulate phased carbon emission reduction plans and phased carbon emission plans, and calculate the planned carbon emissions; S2: Obtain warehouse operation and management data, and calculate the estimated carbon emissions for the current planning cycle; S3: If the estimated carbon emissions exceed the planned carbon emissions within the current planning period, the necessary carbon emissions will be calculated. If the necessary carbon emissions do not exceed the planned carbon emissions, transportation and storage will be carried out according to the optimized plan. Otherwise, emission increase / decrease recommendations will be generated based on the emission increase / decrease optimization strategy.

2. The method for dynamic optimization of warehouse carbon neutrality based on multimodal data fusion according to claim 1, characterized in that, The calculation of necessary carbon emissions includes: obtaining cargo information and cargo storage information; developing multiple simulation schemes based on multiple devices to simulate the transportation status and operation data of each transportation device, and calculating the transportation carbon emissions in each simulation scheme; selecting the simulation scheme with the minimum transportation carbon emissions as the optimized scheme, and the sum of the transportation carbon emissions and the operating carbon emissions corresponding to the optimized scheme is the necessary carbon emissions.

3. A dynamic optimization method for warehouse carbon neutrality based on multimodal data fusion according to claim 1 or 2, characterized in that, The emission reduction optimization strategy includes: acquiring emission reduction measures and calculating the estimated carbon emission reduction, and using the difference between the necessary carbon emissions and the estimated carbon emission reduction as the new emission reduction; acquiring the current emission reduction measures and the preset carbon emission reduction costs corresponding to each emission reduction measure, generating multiple emission reduction measure combinations and calculating the corresponding estimated emission reduction costs; and outputting multiple emission reduction measure combinations and estimated emission reduction costs.

4. The method for dynamic optimization of warehouse carbon neutrality based on multimodal data fusion according to claim 1, characterized in that, This includes transportation equipment anomaly detection: acquiring abnormal task conditions, performing task statistics and judging abnormal task conditions when the time from the start-up to the shutdown of the transportation equipment exceeds a preset first time threshold, and generating an optimized task flow signal if abnormal task conditions are detected.

5. A dynamic optimization method for warehouse carbon neutrality based on multimodal data fusion according to claim 1, 2, or 4, characterized in that, This includes warehouse operation anomaly detection: acquiring abnormal power consumption, and generating an optimized operation power consumption signal if abnormal power consumption is detected. The abnormal power consumption includes being in an abnormal lighting period and the continuous lighting duration exceeds a preset lighting duration threshold, or being in an abnormal power consumption period and the abnormal power duration exceeds a preset power duration threshold.

6. A dynamic optimization method for warehouse carbon neutrality based on multimodal data fusion according to claim 1, 2, or 4, characterized in that, The operational status prediction model is trained, and the operational path and conditions of the transportation equipment are used as outputs to output the warehouse operation data.

7. The method for dynamic optimization of warehouse carbon neutrality based on multimodal data fusion according to claim 3, characterized in that, The emission reduction optimization strategy is configured with an effectiveness correction sub-strategy, which adjusts the corresponding optimization coefficient according to the usage time of the emission reduction measures.

8. The method for dynamic optimization of warehouse carbon neutrality based on multimodal data fusion according to claim 4, characterized in that, The abnormal task conditions include standby time exceeding the second time threshold or idle time exceeding the third time threshold, and the idle time accounting for a preset percentage of the total task time.

9. The method for dynamic optimization of warehouse carbon neutrality based on multimodal data fusion according to claim 6, characterized in that, The operational status prediction model is configured with a model correction strategy. The model correction strategy compares the actual transportation data with the operational data and corrects the operational status prediction model when the difference between the actual transportation data and the operational data is greater than a preset error threshold.

10. A dynamic optimization method for warehouse carbon neutrality based on multimodal data fusion according to claim 2, characterized in that, The operational carbon emissions are obtained by converting energy into carbon from historical data on energy consumption monitoring of building electrical equipment and consumption records of operational materials.

Images