Zero-carbon park intelligent system based on AI large model

Abstract

The invention relates to the technical field of electric energy storage, and discloses a zero-carbon park intelligent system based on an AI large model. The zero-carbon park intelligent system comprises a plurality of new energy power generation device information modules, a time division module, a weather information module, a park prediction weather information data module, an enterprise production information module, a supply pressure analysis module, an abnormal energy consumption evaluation module and an intelligent scheduling module. Through cooperative work of the modules, new energy power generation data, meteorological prediction and actual data and enterprise production power consumption information are integrated, and a high-precision power supply pressure prediction model is constructed; the electric energy supply pressure grade and the enterprise abnormal energy consumption grade of each time period are dynamically evaluated based on historical and real-time data, and the load prediction accuracy and the energy efficiency supervision capability are improved; the power supply pressure is effectively relieved, the resource configuration is optimized, and the on-site consumption of new energy is promoted; and the stability, the economy and the green low-carbon level of the park micro-grid are integrally enhanced.

Description

Technical Field

[0001] This invention relates to the field of energy storage technology, specifically to a zero-carbon intelligent park system based on an AI big data model. Background Technology

[0002] Against the backdrop of the world actively addressing climate change and vigorously promoting the "dual carbon" goal, industrial parks, as major contributors to energy consumption and carbon emissions, urgently need to undergo low-carbon transformation. Zero-carbon parks, as an advanced form of industrial park development, aim to achieve self-balancing of carbon emissions through the integration of clean energy, intelligent technologies, and other means, thus becoming a key force in promoting green development.

[0003] Existing zero-carbon parks utilize spaces such as building rooftops, facades, and carports to install photovoltaic panels, deploy small-scale wind power facilities in open areas or along roads around the park, and use organic waste within the park to generate electricity or heat, thereby reducing dependence on traditional energy sources.

[0004] However, renewable energy generation is greatly affected by weather conditions, and its power generation is intermittent and fluctuates. For example, photovoltaic power generation generates electricity when there is sunlight during the day and stops generating electricity when there is no sunlight at night, and the power generation fluctuates with changes in sunlight intensity. Wind power generation is also affected by wind speed, and unstable wind speed will lead to unstable power generation. This makes it difficult to accurately predict renewable energy generation and make it difficult to plan power supply in advance, resulting in a mismatch between power supply and demand, and potentially leading to power shortages or surpluses. Summary of the Invention

[0005] The purpose of this invention is to provide a zero-carbon intelligent park system based on an AI large-scale model, solving the following technical problems: How to improve the effective utilization rate of electricity in the industrial park.

[0006] The objective of this invention can be achieved through the following technical solutions: A zero-carbon park intelligent system based on an AI large-scale model, the zero-carbon park intelligent system comprising: Several new energy power generation device information modules are used to collect and store power generation information data of each new energy power generation device; The time division module is used to divide 24 hours into several basic time units; The meteorological information module is used to collect forecast meteorological information data and actual meteorological information data in accordance with meteorological collection rules; The park's forecast meteorological information data module is used to analyze the forecast meteorological information data for each basic time unit of the target date, the forecast meteorological information data for each basic time unit within the past preset time period, and the actual meteorological information data to obtain the park's forecast meteorological information data for each basic time unit. The enterprise production information module is used to collect and store production information data of various enterprises in the park, including production volume and electricity consumption. The supply pressure analysis module is used to analyze the predicted meteorological information data of the park and the electricity consumption of each enterprise in the past preset number of days for each basic time unit to obtain the power supply pressure level of each basic time unit. The abnormal energy consumption assessment module is used to analyze the production information data and actual meteorological information data of each enterprise in the park to obtain the abnormal energy consumption level of each enterprise. The intelligent scheduling module is used to determine the scheduling strategy based on the power supply pressure level of each basic time unit and the abnormal energy consumption level of each enterprise.

[0007] As a further aspect of the present invention: the meteorological data collection rules include predictive meteorological information data collection rules and actual meteorological information data collection rules. The predictive meteorological information data collection rules are to collect meteorological forecast information data for each basic time unit of the target date in the area where the park is located at a preset time before the target date.

[0008] As a further aspect of the present invention: the actual meteorological information data collection rules are as follows; S1: Analyze the predicted meteorological information data of the park for each basic time unit of the target date to obtain the predicted meteorological change index for each basic time unit of the target date; S2: Obtain the number of feature points for each basic time unit of the target date based on the predicted meteorological change index for each basic time unit of the target date; S3: The number of feature points in each basic time unit of the target date is used to obtain the collection interval duration of each basic time unit of the target date; S4: The meteorological information module collects actual meteorological information data according to the collection interval of each basic time unit of the target date.

[0009] As a further aspect of the present invention: the park forecast meteorological information data module includes several park image acquisition units, recognition units, and analysis units; the park image acquisition unit is used to acquire image information data of a designated area of ​​the park; the recognition unit is a trained convolutional neural network model used to recognize the image information data of the designated area of ​​the park and obtain the environmental distribution data of the park.

[0010] As a further aspect of the present invention: the analysis unit includes: S10: Use the environmental distribution data of the park, the predicted meteorological information data of each basic time unit in the past preset time period and the actual meteorological information data as training samples, input them into the prediction model for training, and obtain the trained meteorological prediction model of the park. S20: Input the predicted meteorological information data of each basic time unit of the target date into the trained park meteorological prediction model to obtain the predicted meteorological information data of each basic time unit of the target date.

[0011] As a further aspect of the present invention: the process for obtaining the power supply pressure level of each basic time unit is as follows: S100: Based on the analysis of the predicted meteorological information data of the park in each basic time unit, the power generation impact index of each new energy power generation device information module is obtained. S200: Based on the power generation impact index of each new energy power generation device information module, the predicted power generation of each basic time unit is obtained through analysis. S300: Based on the predicted power generation of each basic time unit and the electricity consumption of each enterprise within the past preset number of days, the power supply pressure index of each basic time unit is obtained. S400: The power supply pressure index of each basic time unit is analyzed to obtain the power supply pressure level of each basic time unit.

[0012] As a further aspect of the present invention: the power supply pressure levels include low, medium and high levels.

[0013] As a further aspect of the present invention: the process for determining the abnormal energy consumption level of each enterprise is as follows: S1000: Based on the analysis of electricity consumption, production volume, and actual meteorological information data of each enterprise over the past preset number of days, the abnormal electricity consumption index of each enterprise is obtained daily. S2000: Analyze the abnormal power consumption index of each enterprise to obtain the abnormal energy consumption level of each enterprise.

[0014] As a further aspect of the present invention, the abnormal energy consumption levels of each enterprise include normal, slightly abnormal, and severely abnormal.

[0015] As a further aspect of the present invention: the environmental distribution data includes building distribution, vegetation distribution, and geographical environment.

[0016] The beneficial effects of this invention are: (1) This invention collects and stores power generation information data of each new energy power generation device through a new energy power generation device information module; then, it divides 24 hours into several basic time units through a time division module; then, it collects predicted meteorological information data and actual meteorological information data according to meteorological collection rules through a meteorological information module; then, it analyzes the predicted meteorological information data and actual meteorological information data of each basic time unit within a preset time period through a park predicted meteorological information data module to obtain the park predicted meteorological information data of each basic time unit; then, it collects and stores the production information data of each enterprise in the park through an enterprise production information module, the production information data including production volume and electricity consumption; then, it analyzes the park predicted meteorological information data and the electricity consumption of each enterprise within a preset number of days through a supply pressure analysis module. The system analyzes and obtains the power supply pressure level for each basic time unit. Then, through an abnormal energy consumption assessment module, it analyzes production information data and actual meteorological data from various enterprises within the park to obtain the abnormal energy consumption level for each enterprise. Finally, the intelligent dispatch module determines the dispatch strategy based on the power supply pressure level for each basic time unit and the abnormal energy consumption level for each enterprise. This technical solution not only integrates new energy power generation data, meteorological forecasts and actual data, and enterprise production electricity consumption information to construct a high-precision power supply pressure prediction model, but also dynamically assesses the power supply pressure level and abnormal energy consumption level of enterprises for each time period based on historical and real-time data, improving the accuracy of load forecasting and energy efficiency monitoring capabilities. This effectively alleviates power supply pressure, optimizes resource allocation, and promotes the local consumption of new energy. Overall, it enhances the stability, economy, and green low-carbon level of the park's microgrid. (2) This invention calculates the predicted meteorological change index for each basic time unit of the target date by analyzing the change curves of various meteorological parameters over time. This index comprehensively considers the changes of multiple meteorological parameters at multiple reference points, and can comprehensively and accurately reflect the intensity of meteorological changes within each basic time unit. Based on this index, the number of feature points is further determined, so that the selection of feature points closely follows the actual situation of meteorological changes, avoids blindly selecting feature points, and improves the targeting of data collection. In this embodiment, the collection interval is calculated based on the number of feature points, which improves the efficiency of data collection while ensuring the validity of the data. Compared with a fixed collection frequency, this dynamic adjustment method can reasonably allocate data collection resources according to actual needs, avoid excessive data collection during periods of little meteorological change, and ensure sufficient data collection during critical periods of meteorological change, thereby improving the efficiency of resource utilization. Attached Figure Description

[0017] The invention will now be further described with reference to the accompanying drawings.

[0018] Figure 1 This is a system module framework diagram of one embodiment of the present invention. Detailed Implementation

[0019] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0020] Please see Figure 1 As shown, in one embodiment, a zero-carbon park intelligent system based on an AI large model is provided, applicable to distributed new energy systems. The distributed new energy system includes several new energy power generation devices, and the zero-carbon park intelligent system includes: Several new energy power generation device information modules are used to collect and store power generation information data of each new energy power generation device; Specifically, new energy power generation devices can be photovoltaic power generation modules, wind power generation modules, or other new energy power generation devices; The time division module is used to divide 24 hours into several basic time units; Specifically, each basic time unit has the same duration, which can be 15 minutes or 30 minutes; the setting is based on the experience of the staff and will not be detailed here. The meteorological information module is used to collect forecast meteorological information data and actual meteorological information data in accordance with meteorological collection rules; Specifically, the meteorological data collection rules include rules for collecting forecast meteorological information data and rules for collecting actual meteorological information data. The rules for collecting forecast meteorological information data involve collecting meteorological forecast data for each basic time unit of the target date in the area where the park is located at a preset time (e.g., more than 18 hours) before the target date. This meteorological forecast data can be obtained from the meteorological bureau. This technology is existing technology and will not be described in detail here. The park's forecast meteorological information data module is used to analyze the forecast meteorological information data for each basic time unit of the target date, the forecast meteorological information data for each basic time unit within the past preset time period, and the actual meteorological information data to obtain the park's forecast meteorological information data for each basic time unit. Specifically, since the meteorological bureau obtains forecast meteorological information data for the city where the park is located, not for the park itself, the park's forecast meteorological information data module uses an AI big model to combine forecast meteorological information data and actual meteorological information data for each basic time unit within a preset time period, as well as the park's own geographical environment, surrounding building distribution, vegetation cover and other characteristic factors, to deeply correct and optimize the meteorological bureau's data. Through training with a large amount of historical data, the AI ​​big model can calculate forecast meteorological information data for the park that is more in line with the actual situation of the park based on the forecast meteorological information data for the target date, providing a more reliable basis for subsequent energy management. The enterprise production information module is used to collect and store production information data of various enterprises in the park, including production volume and electricity consumption. Specifically, the production volume and electricity consumption of each enterprise are obtained using existing technology, which will not be detailed here; The supply pressure analysis module is used to analyze the predicted meteorological information data of the park and the electricity consumption of each enterprise in the past preset number of days for each basic time unit to obtain the power supply pressure level of each basic time unit. Specifically, the power supply pressure level is used to assess the matching degree between the power generation and power consumption of new energy power generation devices. When the forecast weather information shows sufficient sunshine and stable wind, if the power consumption of various enterprises is at a low level, then the power supply pressure level of this basic time unit is low, which means that the power generated by the new energy power generation devices can easily meet the power demand of enterprises in the park, and there may even be surplus power that can be stored or transmitted. Conversely, if the forecast weather conditions are unfavorable, such as cloudy days or no wind, resulting in a significant reduction in the power generation of new energy power generation devices, while the power consumption of various enterprises is at its peak, then the power supply pressure level will increase, indicating that the park may face a tight power supply situation, and corresponding measures need to be taken to ensure the stability of the power supply, such as starting energy storage equipment or purchasing power from the external grid. Through accurate assessment of the power supply pressure level, park managers can formulate reasonable power dispatch strategies in advance, optimize energy allocation, improve energy utilization efficiency, reduce electricity costs, and ensure the safe, stable, and economical operation of the park's power system. The abnormal energy consumption assessment module is used to analyze the production information data and actual meteorological information data of each enterprise in the park to obtain the abnormal energy consumption level of each enterprise. Specifically, the abnormal energy consumption level is used to assess the degree of abnormality in the energy consumption of each enterprise in the park. In the actual production process, factors such as the production process, equipment status and production arrangement of different enterprises will affect their energy consumption. The abnormal energy consumption assessment module will take these factors into account and compare and analyze the actual energy consumption of enterprises with the normal energy consumption model built based on their production information data and actual meteorological information data. The intelligent scheduling module is used to determine the scheduling strategy based on the power supply pressure level of each basic time unit and the abnormal energy consumption level of each enterprise. Through the above technical solution, this embodiment first collects and stores power generation information data of each new energy power generation device through the new energy power generation device information module; then, the 24 hours are divided into several basic time units through the time division module; then, the meteorological information module collects predicted meteorological information data and actual meteorological information data according to meteorological collection rules; then, the park predicted meteorological information data module analyzes the predicted meteorological information data and actual meteorological information data of each basic time unit within the past preset time period to obtain the park predicted meteorological information data for each basic time unit; then, the enterprise production information module collects and stores the production information data of each enterprise in the park, including production volume and electricity consumption; finally, the supply pressure analysis module analyzes the park predicted meteorological information data of each basic time unit and the electricity consumption of each enterprise within the past preset number of days. The system analyzes data to obtain the power supply pressure level for each basic time unit. Then, an abnormal energy consumption assessment module analyzes production information data and actual meteorological data from various enterprises within the park to obtain the abnormal energy consumption level for each enterprise. Finally, an intelligent scheduling module determines the scheduling strategy based on the power supply pressure level for each basic time unit and the abnormal energy consumption level for each enterprise. This technical solution not only integrates new energy power generation data, meteorological forecasts and actual data, and enterprise production electricity consumption information to construct a high-precision power supply pressure prediction model, but also dynamically assesses the power supply pressure level and abnormal energy consumption level of enterprises for each time period based on historical and real-time data, improving the accuracy of load forecasting and energy efficiency monitoring capabilities. This effectively alleviates power supply pressure, optimizes resource allocation, and promotes the local consumption of new energy. Overall, it enhances the stability, economy, and green low-carbon level of the park's microgrid.

[0021] As one embodiment of the present invention, the actual meteorological information data collection rules are as follows: S1: Analyze the predicted meteorological information data of the park for each basic time unit of the target date to obtain the predicted meteorological change index for each basic time unit of the target date; Specifically, the meteorological forecast information data includes curves showing the changes of various meteorological parameters over time; Select a sufficient number of reference points on the time-varying curves of various meteorological parameters; Using Formula 1: ; Calculate the predicted weather change index B for the i-th basic time unit of the target date. i ; Where f(X) is the judgment function, f(X)=X when X>0; f(X)=0 when X≤0; M is the number of meteorological parameters, m∈M; θ m Q represents the weighting coefficient for the m-th meteorological parameter; N is the number of reference points, n∈N; mn Q represents the value of the m-th meteorological parameter at the n-th reference point; m (t) represents the curve of the m-th meteorological parameter over time; t i Δt is the start time of the i-th basic time unit; Δt is the duration of the basic time unit. It should be noted that the weighting coefficients of each meteorological parameter and the duration of the basic time unit are preset values, set based on historical data fitting, which are existing technologies and will not be described in detail here.

[0022] S2: Obtain the number of feature points for each basic time unit of the target date based on the predicted meteorological change index for each basic time unit of the target date; Specifically, through Formula 2: ; Calculate the number H of feature points in the i-th basic time unit of the target date; Where H0 is the number of basic feature points; Z0 is the preset constant for feature points; To Round up; It should be noted that the number of basic feature points H0 and the preset constant Z0 of feature points are set based on historical data fitting, which is existing technology and will not be described in detail here.

[0023] S3: The number of feature points in each basic time unit of the target date is used to obtain the collection interval duration of each basic time unit of the target date; Specifically, through Formula 3 ; Calculate the acquisition interval tc of the i-th basic time unit of the target date. i ; S4: The meteorological information module collects actual meteorological information data according to the collection interval of each basic time unit of the target date; Through the above technical solution, this embodiment calculates the predicted meteorological change index for each basic time unit of the target date by analyzing the change curves of various meteorological parameters over time. This index comprehensively considers the changes of multiple meteorological parameters at multiple reference points, and can comprehensively and accurately reflect the intensity of meteorological changes within each basic time unit. Based on this index, the number of feature points is further determined, ensuring that the selection of feature points closely follows the actual situation of meteorological changes, avoiding blind selection of feature points, and improving the targeting of data collection. This embodiment calculates the collection interval based on the number of feature points, ensuring that the collection interval is shorter during periods of intense meteorological changes, allowing for more intensive data collection and capturing the details of meteorological changes; while the collection interval is longer during periods of relatively stable meteorological changes, reducing unnecessary data collection, and improving the efficiency of data collection while ensuring data validity. Compared with a fixed collection frequency, this dynamic adjustment method can reasonably allocate data collection resources according to actual needs, avoiding excessive data collection during periods of little meteorological change and thus avoiding resource waste, while ensuring sufficient data collection during critical periods of meteorological change, thereby improving resource utilization efficiency.

[0024] As one embodiment of the present invention, the park forecast meteorological information data module includes several park image acquisition units, recognition units, and analysis units; the park image acquisition unit is used to acquire image information data of a designated area of ​​the park; the recognition unit is a trained convolutional neural network model, used to recognize the image information data of the designated area of ​​the park and obtain the environmental distribution data of the park; Specifically, environmental distribution data can include building distribution, vegetation distribution, and geographical environment; the designated area of ​​the park refers to the relevant internal and external areas of the park that need to be involved in the process of identifying the environmental distribution data of the park. Specifically, the training process of the convolutional neural network model is an existing technology and will not be described in detail here; Through the above technical solution, this embodiment can accurately collect image information of a designated area in the park through the park image acquisition unit, providing rich basic data for subsequent analysis; the convolutional neural network model, as the recognition unit, has been trained with a large amount of data and has powerful image recognition capabilities, which can efficiently and accurately process the collected image information to obtain environmental distribution data of the park, including the distribution of buildings, vegetation, and geographical environment. This combination avoids the errors and omissions that may occur in manual recognition, and can also update the layout of the park in a timely manner, greatly improving the accuracy and efficiency of data acquisition. It provides a more comprehensive and accurate basis for the park's characteristic factors for the park's meteorological information data module, which helps the AI ​​big model to more accurately correct and optimize the meteorological bureau data, thereby improving the performance and reliability of the entire zero-carbon park intelligent system.

[0025] In one embodiment of the present invention, the analysis unit includes: S10: Use the environmental distribution data of the park, the predicted meteorological information data of each basic time unit in the past preset time period and the actual meteorological information data as training samples, input them into the prediction model for training, and obtain the trained meteorological prediction model of the park. Specifically, the training process of the park's weather forecasting model is based on existing technology and will not be described in detail here; S20: Input the predicted meteorological information data of each basic time unit of the target date into the trained park meteorological prediction model to obtain the predicted meteorological information data of the park for each basic time unit of the target date. Through the above technical solution, in this embodiment, the analysis unit uses the building distribution, vegetation distribution, geographical environment, and predicted and actual meteorological information data of each basic time unit within the past preset time period as training samples to train the park's meteorological prediction model, making the trained prediction model more in line with the actual situation of the park; then, by inputting the predicted meteorological information data of the target date, the predicted meteorological information data of the park can be accurately obtained; this effectively overcomes the deficiency of insufficient targeting of meteorological bureau data, provides a more reliable basis for energy management, helps to improve the energy utilization efficiency of the park, and ensures the stable operation of the power system.

[0026] As one embodiment of the present invention, the process for obtaining the power supply pressure level of each basic time unit is as follows: S100: Based on the analysis of the predicted meteorological information data of the park in each basic time unit, the power generation impact index of each new energy power generation device information module is obtained. Specifically, through Formula 4: ; Calculate the power generation impact index F of the j-th new energy power generation device information module in the i-th basic time unit. ij ; in, F is the average value of the predicted meteorological information data of the park for the m-th meteorological parameter in the i-th basic time unit of the target date; m0 μ is the preset value for the m-th meteorological parameter; mj ρ is the correlation coefficient between the m-th meteorological parameter and the j-th new energy power generation device; mj C represents the preset weighting coefficient of the m-th meteorological parameter for the j-th new energy power generation device; m This is a preset constant for the m-th meteorological parameter; It should be noted that when the m-th meteorological parameter is positively correlated with the power generation of the j-th new energy power generation device, the correlation coefficient is 1; when the m-th meteorological parameter is negatively correlated with the power generation of the j-th new energy power generation device, the correlation coefficient is -1. It should be noted that the preset values ​​of each meteorological parameter, the preset weighting coefficients of each meteorological parameter for each new energy power generation device, and the preset constants of each meteorological parameter are all preset values, set based on historical data fitting, and are existing technologies, which will not be described in detail here.

[0027] S200: Based on the power generation impact index of each new energy power generation device information module, the predicted power generation of each basic time unit is obtained through analysis. Specifically, through Formula 5: ; Calculate the predicted power generation W for the i-th basic time unit. i ; Where J represents the number of new energy power generation devices, j∈J; W j0 C represents the preset power generation of the j-th new energy power generation device under preset environmental conditions. j Let j be the preset constant for the power generation of the j-th new energy power generation device; It should be noted that the preset environmental conditions refer to the state of each meteorological parameter at preset values; It should be noted that the preset power generation constants and preset power generation of each new energy power generation device under preset environmental conditions are preset values, set based on historical data fitting, and are existing technologies, which will not be described in detail here.

[0028] S300: Based on the predicted power generation of each basic time unit and the electricity consumption of each enterprise within the past preset number of days, the power supply pressure index of each basic time unit is obtained. Specifically, through Formula Six: ; Calculate the power supply pressure index T for the i-th basic time unit. i ; Where K is the preset number of past days, k∈K; U ki ε represents the cumulative electricity consumption of each enterprise over the past k-th day in the i-th basic time unit; k The weighting coefficient for day k; It should be noted that the weighting coefficient ε on day k k These are preset values, set based on historical data fitting, and are existing technologies, which will not be described in detail here.

[0029] S400: Analyze the power supply pressure index of each basic time unit to obtain the power supply pressure level of each basic time unit; Specifically, the power supply pressure index T of the i-th basic time unit is... i Compare with preset thresholds [Z1, Z2]; When Ti When Z1 ≤ Z1, the power supply pressure level of the i-th basic time unit is high level; When Z1 < T i When Z2 ≤ Z2, the power supply pressure level of the i-th basic time unit is medium level; When Z2 < T i At that time, the power supply pressure level of the i-th basic time unit is low. It should be noted that Z1∈[1.1, 1.2], Z1<Z2; the specific values ​​of the preset threshold [Z1, Z2] are set based on historical data fitting, which is existing technology and will not be described in detail here; Through the above technical solution, this embodiment calculates the power generation impact index by analyzing the predicted meteorological information data of the park, which can accurately grasp the role of meteorology in new energy power generation devices and provide a reliable basis for subsequent power generation forecasting. The power generation forecast is calculated based on the power generation impact index, making the forecast results more in line with reality. The power supply pressure index is calculated by introducing the electricity consumption of enterprises in the past preset number of days and the weighting coefficient, which comprehensively considers the historical patterns and recent trends of electricity demand and accurately reflects the supply and demand balance. Finally, the pressure index is compared with the preset threshold to determine the level, and the power supply pressure status at different times is presented in an intuitive way. This helps park managers to formulate targeted strategies in advance, such as increasing power security measures when the level is high and reasonably arranging power reserves or external transmission when the level is low, effectively improving the scientificity and flexibility of power supply management and ensuring the stable operation of the park's power system.

[0030] As one embodiment of the present invention, the process for determining the abnormal energy consumption level of each enterprise is as follows: S1000: Based on the analysis of electricity consumption, production volume, and actual meteorological information data of each enterprise over the past preset number of days, the abnormal electricity consumption index of each enterprise is obtained daily. Specifically, through Formula Seven: ; Calculate the abnormal power consumption index of the s-th enterprise. ; Among them, Y sk X represents the projected electricity consumption of the s-th enterprise on day k; sk Let 's' be the production volume of the 's'th enterprise on day k. ω represents the electricity consumption of each device under normal operating conditions and preset environmental conditions for the production output of the s-th enterprise on day k; ms Let m be the weight of the impact of the m-th meteorological parameter on the s-th enterprise's electricity consumption; This represents the average value of the m-th meteorological parameter on the k-th day. Let be the electricity consumption of the s-th enterprise on the k-th day; It should be noted that the production volume of each enterprise on day k, the electricity consumption of each piece of equipment under the preset environmental conditions, and the weights of the impact of each meteorological parameter on the electricity consumption of each enterprise are all preset values, set based on historical data fitting, which is existing technology and will not be described in detail here.

[0031] S2000: Analyze the abnormal power consumption index of each enterprise to obtain the abnormal energy consumption level of each enterprise; Specifically, the abnormal power consumption index of the s-th enterprise on the k-th day... Compare with preset comparison values ​​[E1, E2]; When 0 < When E1 is less than or equal to 1, the abnormal power consumption level of the s-th enterprise on the k-th day is normal. When E1 < When E2 is less than or equal to 2, the abnormal power consumption level of the s-th enterprise on day k is classified as slightly abnormal. When E2 < At that time, the abnormal power consumption level of the s-th enterprise on the k-th day was classified as severely abnormal; It should be noted that the preset comparison values ​​[E1, E2] are preset values, set based on historical data fitting, and are existing technologies, which will not be described in detail here; Through the above technical solution, this embodiment can accurately calculate the daily abnormal power consumption index by analyzing the electricity consumption, production volume and meteorological information data of the past preset number of days. It takes into account the influence of multiple factors such as production and weather, making the results more scientific and comprehensive. Based on the preset comparison value, the index is converted into normal, slightly abnormal and severely abnormal levels, which can provide a reliable basis for the intelligent scheduling module, help to reasonably adjust the power consumption of enterprises, and ensure the stable power supply and efficient utilization of the park.

[0032] As one embodiment of the present invention, the scheduling strategy of the intelligent scheduling module includes: When the power supply pressure level of any basic time unit is high: The system proactively notifies enterprises with minor or severe power consumption anomalies, reminding them to adjust production plans, optimize equipment usage, and reduce energy consumption. Within this basic time unit, the intelligent dispatch module prioritizes reducing the power load of non-critical production processes for enterprises with minor or severe anomalies, thereby reducing overall power demand in the park. Simultaneously, based on the power status of the park's energy storage devices, if sufficient power is available, the devices are activated to supply power to the park, alleviating power supply pressure. If power is insufficient, the system immediately contacts the external power grid and purchases an appropriate amount of electricity according to pre-set power purchase agreements and prices to ensure normal power supply for key enterprises within the park (such as those involved in public welfare, high-tech production, etc.). Furthermore, the intelligent dispatch module monitors the power generation of new energy power generation devices in real time, and promptly adjusts power rationing measures and power purchase plans should power generation increase. When the power supply pressure level of any basic time unit is medium: The intelligent dispatch module first issues energy consumption warnings to enterprises with severely abnormal energy consumption levels, adjusts production plans, and assesses the energy storage equipment in the park. If the energy storage equipment has a moderate level of power, the charging power of the energy storage equipment can be appropriately reduced, and some of the power can be used for power supply in the park. In addition, the intelligent dispatch module closely monitors the power generation trend of new energy power generation devices and changes in the electricity demand of enterprises, and flexibly adjusts the power supply strategy according to the actual situation to ensure the stability of the park's power supply.

[0033] When the power supply pressure level of each basic time unit is low: If the battery storage devices in the park are not fully charged, the intelligent dispatch module will prioritize using the surplus electricity generated by the new energy power generation devices to charge the battery storage devices, thereby improving the park's power reserve capacity to cope with potential future power supply pressure. For enterprises with normal abnormal energy consumption levels and large production volumes, they are encouraged to appropriately increase their production load without affecting their normal production, making full use of the park's abundant power resources and improving production efficiency. At the same time, the intelligent dispatch module will continuously monitor the power generation of the new energy power generation devices. If the power generation continues to exceed the park's power demand, some of the electricity may be sent to surrounding areas with demand, achieving reasonable allocation and optimized utilization of energy.

[0034] Through the above technical solution, in this embodiment, the intelligent scheduling module will feed back the scheduling information to the park management personnel in real time during the entire scheduling process, including the list of enterprises subject to power restrictions, the amount of electricity purchased, and the status of energy storage equipment, so that the management personnel can understand the power supply situation in the park in a timely manner and make scientific and reasonable decisions.

[0035] The foregoing has provided a detailed description of one embodiment of the present invention, but this description is merely a preferred embodiment and should not be construed as limiting the scope of the invention. All equivalent variations and modifications made within the scope of the claims of this invention should still fall within the patent coverage of this invention.

Claims

1. A zero-carbon industrial park intelligent system based on an AI large-scale model, characterized in that, The intelligent system for the zero-carbon industrial park includes: Several new energy power generation device information modules are used to collect and store power generation information data of each new energy power generation device; The time division module is used to divide 24 hours into several basic time units; The meteorological information module is used to collect forecast meteorological information data and actual meteorological information data in accordance with meteorological collection rules; The park's forecast meteorological information data module is used to analyze the forecast meteorological information data for each basic time unit of the target date, the forecast meteorological information data for each basic time unit within the past preset time period, and the actual meteorological information data to obtain the park's forecast meteorological information data for each basic time unit. The enterprise production information module is used to collect and store production information data of various enterprises in the park, including production volume and electricity consumption. The supply pressure analysis module is used to analyze the predicted meteorological information data of the park and the electricity consumption of each enterprise in the past preset number of days for each basic time unit to obtain the power supply pressure level of each basic time unit. The abnormal energy consumption assessment module is used to analyze the production information data and actual meteorological information data of each enterprise in the park to obtain the abnormal energy consumption level of each enterprise. The intelligent scheduling module is used to determine the scheduling strategy based on the power supply pressure level of each basic time unit and the abnormal energy consumption level of each enterprise.

2. The intelligent zero-carbon park system based on an AI large-scale model according to claim 1, characterized in that, The meteorological data collection rules include forecast meteorological information data collection rules and actual meteorological information data collection rules. The forecast meteorological information data collection rules are to collect meteorological forecast information data for each basic time unit of the target date in the park area at a preset time before the target date.

3. The intelligent zero-carbon park system based on an AI large-scale model according to claim 2, characterized in that, The actual rules for collecting meteorological information data are as follows: S1: Analyze the predicted meteorological information data of the park for each basic time unit of the target date to obtain the predicted meteorological change index for each basic time unit of the target date; S2: Obtain the number of feature points for each basic time unit of the target date based on the predicted meteorological change index for each basic time unit of the target date; S3: The number of feature points in each basic time unit of the target date is used to obtain the collection interval duration of each basic time unit of the target date; S4: The meteorological information module collects actual meteorological information data according to the collection interval of each basic time unit of the target date.

4. The intelligent zero-carbon park system based on an AI large-scale model according to claim 3, characterized in that, The park's predicted meteorological information data module includes several park image acquisition units, recognition units, and analysis units; the park image acquisition units are used to acquire image information data of a designated area in the park; the recognition units are trained convolutional neural network models used to recognize the image information data of the designated area in the park and obtain the park's environmental distribution data.

5. The intelligent zero-carbon park system based on an AI large-scale model according to claim 4, characterized in that, The analysis unit includes: S10: Use the environmental distribution data of the park, the predicted meteorological information data of each basic time unit in the past preset time period and the actual meteorological information data as training samples, input them into the prediction model for training, and obtain the trained meteorological prediction model of the park. S20: Input the predicted meteorological information data of each basic time unit of the target date into the trained park meteorological prediction model to obtain the predicted meteorological information data of each basic time unit of the target date.

6. The intelligent zero-carbon park system based on an AI large-scale model according to claim 5, characterized in that, The process for obtaining the power supply pressure level for each basic time unit is as follows: S100: Based on the analysis of the predicted meteorological information data of the park in each basic time unit, the power generation impact index of each new energy power generation device information module is obtained. S200: Based on the power generation impact index of each new energy power generation device information module, the predicted power generation of each basic time unit is obtained through analysis. S300: Based on the predicted power generation of each basic time unit and the electricity consumption of each enterprise within the past preset number of days, the power supply pressure index of each basic time unit is obtained. S400: The power supply pressure index of each basic time unit is analyzed to obtain the power supply pressure level of each basic time unit.

7. The intelligent zero-carbon park system based on an AI large-scale model according to claim 6, characterized in that, The power supply pressure levels include low, medium and high levels.

8. The intelligent zero-carbon park system based on an AI large-scale model according to claim 7, characterized in that, The process for determining the abnormal energy consumption levels of each enterprise is as follows: S1000: Based on the analysis of electricity consumption, production volume, and actual meteorological information data of each enterprise over the past preset number of days, the abnormal electricity consumption index of each enterprise is obtained daily. S2000: Analyze the abnormal power consumption index of each enterprise to obtain the abnormal energy consumption level of each enterprise.

9. A zero-carbon intelligent park system based on an AI large model as described in claim 8, characterized in that... The abnormal energy consumption levels of each enterprise are categorized as normal, slightly abnormal, and severely abnormal.

10. A zero-carbon industrial park intelligent system based on an AI large-scale model according to claim 9, characterized in that, The environmental distribution data includes building distribution, vegetation distribution, and geographical environment.

Images