A source-load dynamic matching building flexible intelligent control system

Abstract

The application provides a building flexible intelligent control system for source-load dynamic matching, and relates to the technical field of building source-load-storage collaborative power supply and distribution control.The building comfort degree, photovoltaic output, grid-connected electricity, load, energy storage and distribution branch bearing data are collected, the source-load topology matching state, energy storage compartment scheduling demand and source-load fluctuation suppression state are evaluated, and the photovoltaic power supply path control, energy storage charging and discharging, key load power supply guarantee, flexible load limiting and slow variation control instructions are generated, so that the dynamic collaboration among the building source, load and storage is realized, and the photovoltaic consumption capacity, peak load shifting capacity, key load power supply guarantee capacity and grid-connected operation stability are improved.

Description

Technical Field

[0001] This invention relates to the field of building power supply and distribution control technology, specifically a building flexible intelligent control system for dynamic matching of power sources and loads. Background Technology

[0002] Against the backdrop of new power system construction and the development of low-carbon building energy consumption, building internal power systems are gradually shifting from a single grid power supply mode to a composite power supply and distribution mode that integrates photovoltaic power generation, energy storage devices, flexible loads, and grid demand response. In energy consumption scenarios such as office buildings and commercial buildings, rooftop photovoltaics, facade photovoltaics, energy storage batteries, central air conditioning, fresh air systems, lighting systems, and office equipment are all connected to the building power distribution system, transforming buildings from traditional power terminals into energy-consuming units with source, load, and storage synergy.

[0003] Existing building energy control methods primarily rely on timed control, zoned control, or single-equipment energy-saving control. These methods typically control the operation of air conditioning, lighting, and energy storage equipment based on fixed time periods, fixed temperature thresholds, fixed lighting strategies, or peak-valley electricity pricing rules. While these methods can reduce energy consumption of local equipment to some extent, their control focus is mainly on the load side. They lack a comprehensive assessment of photovoltaic output, energy storage capacity, distribution branch capacity, and grid connection electrical status, making it difficult to achieve dynamic matching control for the overall building power supply and distribution status.

[0004] With the increasing proportion of building-integrated photovoltaics (BIPV), PV output becomes volatile due to weather, irradiance, temperature, and shading conditions, while building loads are randomized by office hours, occupancy density, meetings, indoor environmental requirements, and equipment start-up / shutdown status. It is difficult to synchronize source-side power supply capacity with load-side electricity demand, leading to problems such as insufficient energy storage utilization during PV surplus periods, untimely energy storage support during peak load periods, and non-critical loads occupying critical load power margins. Existing systems typically only control charging and discharging based on energy storage state of charge or peak-valley electricity prices, failing to finely allocate energy storage capacity according to PV surplus power, peak shaving needs, comfort assurance needs, and critical load supply requirements. This results in insufficient synergy between energy storage systems and PV integration, peak shaving, critical load supply assurance, and comfort assurance.

[0005] The load-bearing status of internal building power distribution branches, power fluctuations at the grid connection point, and voltage deviations at the grid connection point also affect the stability of building power supply. Existing building energy consumption control systems typically focus on energy consumption statistics and equipment start-up and shutdown control, lacking a coordinated assessment of changes in active power at the grid connection point, photovoltaic output, building load changes, and voltage deviations. This makes it difficult to promptly control energy storage converters for short-term power compensation or voltage support, and also makes it difficult to limit the power increase rate of flexible loads such as air conditioning, fresh air, and lighting during sudden load surges, easily causing source-load fluctuations to be transmitted to the building's grid connection point.

[0006] Existing building flexible control technologies still suffer from insufficient coordination between energy-saving goals and human comfort. Simply reducing the power of air conditioning, fresh air, or lighting may cause indoor temperature, illuminance, or carbon dioxide concentration to deviate from the comfortable range; on the other hand, prioritizing comfort may weaken demand response and peak shaving effects. Summary of the Invention

[0007] The purpose of this invention is to provide a flexible intelligent control system for dynamic matching of source and load in buildings, so as to solve the problems mentioned in the background art.

[0008] To achieve the above objectives, the present invention provides the following technical solution: A flexible intelligent control system for dynamic matching of source and load in buildings, specifically comprising: The integrated data acquisition module is used to collect building indoor comfort constraint data, source-side and grid-connected electrical data, load-side load data, energy storage status and power data, and distribution branch load data; The source-load topology matching assessment module is used to calculate the source-load power balance coefficient, the remaining load factor of the distribution branch, the energy storage support coefficient and the flexible load adjustment coefficient, and comprehensively calculate the source-load topology matching index to determine the source-load topology matching status of the building, generate the first assessment result, and output the current topology maintenance strategy, the local power distribution reconfiguration strategy or the forced source-load matching strategy. The energy storage virtual compartment scheduling module is used to calculate the surplus photovoltaic power, peak reduction demand power, comfort deviation coefficient and comfort guarantee demand power, and comprehensively calculate the energy storage virtual compartment scheduling coefficient, determine the compartment scheduling demand status of the energy storage system, generate the second evaluation result, and output the first compartment scheduling strategy, the second compartment scheduling strategy or the third compartment scheduling strategy. The source load fluctuation suppression assessment module is used to calculate the grid connection point power fluctuation coefficient, grid connection point voltage deviation coefficient, photovoltaic output change coefficient and building load change coefficient, and comprehensively calculate the source load fluctuation suppression coefficient, determine the building source load fluctuation suppression status, generate the third assessment result, and output the fluctuation allowance strategy, energy storage short-term compensation strategy or voltage support and load gradual change strategy. The flexible control strategy generation module receives the first assessment result, the second assessment result, and the third assessment result, performs comprehensive strategy determination and control command conflict handling, generates the final execution strategy, and outputs the final building flexible control command.

[0009] Furthermore, the integrated data acquisition module includes an indoor comfort constraint acquisition unit, a source-side and grid-connected electrical acquisition unit, a load-side load acquisition and reduction calculation unit, an energy storage status and power acquisition unit, and a distribution branch load acquisition unit; The indoor comfort constraint acquisition unit is used to collect data on the number of people, indoor temperature, indoor illuminance, and carbon dioxide concentration. The source-side and grid-connected electrical acquisition unit is used to collect the real-time output power of building photovoltaics, grid connection point voltage, grid connection point active power, photovoltaic predicted output power, grid connection point rated voltage, and grid demand response reduction power. The load-side load acquisition and reduction calculation unit is used to acquire the building's total power consumption, target load power, critical load power, and flexible load power, and to calculate the power that can be reduced from the flexible load. The energy storage status and power acquisition unit is used to acquire the energy storage state of charge, energy storage health status, energy storage rated capacity, energy storage current charging power, energy storage current discharging power and energy storage maximum allowable discharge power. The power distribution branch carrying acquisition unit is used to collect the real-time current of each power distribution branch and obtain the rated current of each power distribution branch.

[0010] Furthermore, the source-load topology matching evaluation module includes a source-load power balance calculation unit, a distribution branch load calculation unit, an energy storage support capacity calculation unit, a flexible load adjustment calculation unit, a source-load topology matching index calculation unit, and a first analysis unit; The source-load power balance calculation unit is used to calculate the source-load power balance coefficient based on the building's total power consumption, the real-time output power of the building photovoltaic system, the maximum allowable discharge power of the energy storage system, and the power reduction power of the grid demand response. The power distribution branch load-bearing calculation unit is used to calculate the remaining load-bearing coefficient of each power distribution branch based on the real-time current and rated current of each power distribution branch. The energy storage support capacity calculation unit is used to calculate the available energy storage capacity and energy storage support coefficient based on the energy storage state of charge, energy storage health status and energy storage rated capacity. The flexible load adjustment calculation unit is used to calculate the flexible load adjustment coefficient based on the power reduction capacity of the flexible load and the power of the flexible load.

[0011] Furthermore, the source-load topology matching index calculation unit is used to calculate the source-load topology matching index based on the source-load power balance coefficient, the remaining carrying capacity coefficient of the distribution branch, the energy storage support coefficient, and the flexible load adjustment coefficient. The first analysis unit is used to compare and analyze the source-load topology matching index with a preset first matching threshold and a second matching threshold to obtain a first evaluation result; when the source-load topology matching index reaches the first matching threshold, it is determined that the building source-load topology is in a matching state and a current topology maintenance strategy is generated; when the source-load topology matching index is between the second matching threshold and the first matching threshold, it is determined that the building source-load topology is in a first-level mismatch state and a local power distribution reconfiguration strategy is generated; when the source-load topology matching index is lower than the second matching threshold, it is determined that the building source-load topology is in a second-level mismatch state and a forced source-load matching strategy is generated.

[0012] Furthermore, the energy storage virtual sub-slot scheduling module includes a photovoltaic surplus power calculation unit, a peak reduction and supply guarantee demand calculation unit, a comfort guarantee demand calculation unit, an energy storage virtual sub-slot scheduling coefficient calculation unit, and a second analysis unit; The photovoltaic surplus power calculation unit is used to calculate the photovoltaic surplus power based on the real-time output power of the building photovoltaic system and the total power consumption of the building. The peak reduction and supply guarantee demand calculation unit is used to calculate the peak reduction demand power based on the power reduction power of the power grid demand response, the total power consumption of the building, and the target load power of the building; The comfort guarantee requirement calculation unit is used to calculate the comfort deviation coefficient and comfort guarantee requirement power based on indoor temperature, indoor illuminance, carbon dioxide concentration, number of people, and flexible load power.

[0013] Furthermore, the energy storage virtual compartment scheduling coefficient calculation unit is used to calculate the energy storage virtual compartment scheduling coefficient based on the photovoltaic surplus power, peak reduction demand power, comfort guarantee demand power, critical load power and total building power consumption; The second analysis unit is used to compare and analyze the energy storage virtual sub-slot scheduling coefficient with the preset first sub-slot threshold and second sub-slot threshold to obtain a second evaluation result. When the energy storage virtual sub-slot scheduling coefficient reaches the first sub-slot threshold, it is determined that the energy storage system is in a first-level scheduling demand state, and a first sub-slot scheduling strategy is generated. When the energy storage virtual sub-slot scheduling coefficient is between the second sub-slot threshold and the first sub-slot threshold, it is determined that the energy storage system is in a second-level scheduling demand state, and a second sub-slot scheduling strategy is generated. When the energy storage virtual sub-slot scheduling coefficient is lower than the second sub-slot threshold, it is determined that the energy storage system is in a third-level scheduling demand state, and a third sub-slot scheduling strategy is generated.

[0014] Furthermore, the source-load fluctuation suppression assessment module includes a grid connection point power fluctuation calculation unit, a grid connection point voltage offset calculation unit, a photovoltaic output change calculation unit, a building load change calculation unit, a source-load fluctuation suppression coefficient calculation unit, and a third analysis unit; The grid connection point power fluctuation calculation unit is used to calculate the grid connection point power fluctuation coefficient based on the grid connection point active power at the current sampling time, the grid connection point active power at the previous sampling time, and the total building power consumption. The grid connection point voltage offset calculation unit is used to calculate the grid connection point voltage offset coefficient based on the grid connection point voltage and the grid connection point rated voltage. The photovoltaic output change calculation unit is used to calculate the photovoltaic output change coefficient based on the real-time output power of the building photovoltaic system at the current sampling time and the output power of the building photovoltaic system at the previous sampling time. The building load change calculation unit is used to calculate the building load change coefficient based on the total building power consumption at the current sampling time and the total building power consumption at the previous sampling time.

[0015] Furthermore, the source-load fluctuation suppression coefficient calculation unit is used to calculate the source-load fluctuation suppression coefficient based on the grid connection point power fluctuation coefficient, grid connection point voltage deviation coefficient, photovoltaic output variation coefficient, and building load variation coefficient.

[0016] Furthermore, the third analysis unit is used to compare and analyze the source load fluctuation suppression coefficient with the preset first fluctuation threshold and second fluctuation threshold to obtain a third evaluation result; when the source load fluctuation suppression coefficient is lower than the first fluctuation threshold, it is determined that the building source load fluctuation is in an allowable state and a fluctuation allowable strategy is generated; when the source load fluctuation suppression coefficient is between the first fluctuation threshold and the second fluctuation threshold, it is determined that the building source load fluctuation is in a first-level suppression state and a short-term energy storage compensation strategy is generated; when the source load fluctuation suppression coefficient reaches the second fluctuation threshold, it is determined that the building source load fluctuation is in a second-level suppression state and a voltage support and load gradual change strategy is generated.

[0017] Furthermore, the flexible control strategy generation module includes a comprehensive strategy determination unit and a final control unit; The comprehensive strategy determination unit is used to generate a final execution strategy based on the first evaluation result, the second evaluation result, and the third evaluation result. When the third evaluation result is a level 2 suppression state, the voltage support and load gradual change strategy is set as the priority execution strategy. When the first evaluation result is a level 2 mismatch state, the forced source-load matching strategy is set as the priority execution strategy. When the second evaluation result is a level 1 scheduling demand state, the first sub-compartment scheduling strategy is set as the priority execution strategy. When the third assessment result is an allowable state, the first assessment result is a matched state, and the second assessment result is a level 3 scheduling demand state, the current operation and maintenance strategy is generated; conflict handling is performed on energy storage charging commands, energy storage discharging commands, energy storage converter voltage support commands, flexible load limiting commands, and flexible load gradual change commands. The final control unit is used to generate final building flexible control instructions based on the conflict handling results, including photovoltaic power supply path control instructions, energy storage compartment capacity call instructions, energy storage charge and discharge control instructions, energy storage converter voltage support instructions, critical load supply guarantee instructions, flexible load limiting instructions, flexible load gradual change instructions, and non-critical load load reduction instructions; and is connected to the building power distribution switchgear, energy storage converter, air conditioning controller, fresh air controller, lighting controller, and sub-item electrical equipment controller respectively to execute the final building flexible control instructions.

[0018] Compared with the prior art, the beneficial effects of the present invention are: This invention collects data from the building's source side, load side, energy storage side, power distribution side, and comfort constraints to determine the building's current power supply capacity, electricity demand, energy storage support capacity, and power distribution branch carrying capacity. Based on the determination results, it formulates strategies such as topology preservation, partial power distribution reconfiguration, or forced source-load matching, enabling the power supply relationship between photovoltaic branches, energy storage branches, critical load branches, and flexible load branches to be adjusted according to changes in source-load status. This improves the building's on-site photovoltaic absorption capacity, peak shaving and valley filling capacity, and critical load power supply guarantee capacity.

[0019] This invention also collects data from the building's source side, load side, energy storage side, power distribution side, and comfort constraints to determine the current building's power supply capacity, electricity demand, energy storage support capacity, and power distribution branch carrying capacity. Based on the determination results, it formulates topology preservation, partial power distribution reconfiguration, or forced source-load matching strategies, enabling the power supply relationship between photovoltaic branches, energy storage branches, critical load branches, and flexible load branches to be adjusted according to changes in source-load status, thereby improving the building's on-site photovoltaic absorption capacity, peak shaving and valley filling capacity, and critical load power supply guarantee capacity. Attached Figure Description

[0020] Figure 1 This is a schematic diagram of the overall system flow of the present invention. Detailed Implementation

[0021] To make the above-mentioned objects, features and advantages of the present invention more apparent and understandable, the specific embodiments of the present invention will be described in detail below with reference to the accompanying drawings.

[0022] Many specific details are set forth in the following description in order to provide a full understanding of the invention. However, the invention may also be practiced in other ways different from those described herein, and those skilled in the art can make similar extensions without departing from the spirit of the invention. Therefore, the invention is not limited to the specific embodiments disclosed below.

[0023] Example 1: Please see Figure 1 This invention provides a technical solution: a flexible intelligent control system for dynamic matching of source and load in buildings, specifically comprising: The integrated data acquisition module is used to collect building indoor comfort constraint data, source-side and grid-connected electrical data, load-side load data, energy storage status and power data, and distribution branch load data; The source-load topology matching assessment module is used to calculate the source-load power balance coefficient, the remaining load factor of the distribution branch, the energy storage support coefficient and the flexible load adjustment coefficient, and comprehensively calculate the source-load topology matching index to determine the source-load topology matching status of the building, generate the first assessment result, and output the current topology maintenance strategy, the local power distribution reconfiguration strategy or the forced source-load matching strategy. The energy storage virtual compartment scheduling module is used to calculate the surplus photovoltaic power, peak reduction demand power, comfort deviation coefficient and comfort guarantee demand power, and comprehensively calculate the energy storage virtual compartment scheduling coefficient, determine the compartment scheduling demand status of the energy storage system, generate the second evaluation result, and output the first compartment scheduling strategy, the second compartment scheduling strategy or the third compartment scheduling strategy. The source load fluctuation suppression assessment module is used to calculate the grid connection point power fluctuation coefficient, grid connection point voltage deviation coefficient, photovoltaic output change coefficient and building load change coefficient, and comprehensively calculate the source load fluctuation suppression coefficient, determine the building source load fluctuation suppression status, generate the third assessment result, and output the fluctuation allowance strategy, energy storage short-term compensation strategy or voltage support and load gradual change strategy. The flexible control strategy generation module receives the first assessment result, the second assessment result, and the third assessment result, performs comprehensive strategy determination and control command conflict handling, generates the final execution strategy, and outputs the final building flexible control command.

[0024] In this embodiment, by integrating comprehensive data acquisition, source-load topology matching assessment, energy storage virtual compartment scheduling, source-load fluctuation suppression assessment, and flexible control strategy generation, the building is no longer controlled solely based on a single load or fixed-time strategy. Instead, it can simultaneously consider photovoltaic output, building load, energy storage capacity, grid connection status, and comfort constraints, thereby improving the building's dynamic source-load matching capability and overall flexible control capability.

[0025] Example 2 Please see Figure 1 In the explanation of Embodiment 1, the integrated data acquisition module specifically includes an indoor comfort constraint acquisition unit, a source-side and grid-connected electrical acquisition unit, a load-side load acquisition and reduction calculation unit, an energy storage status and power acquisition unit, and a distribution branch load acquisition unit. The indoor comfort constraint acquisition unit is used to monitor the occupancy status of people in various functional areas of the building in real time. Space occupancy sensors are installed on the ceilings of office areas, meeting rooms, corridors, and public activity areas to collect the number of people Nz(t); the thermal environment status of various functional areas of the building is monitored in real time by installing temperature and humidity sensors at the return air height on the walls of office areas, meeting rooms, and public activity areas to collect indoor temperature Tz(t); the lighting environment status of various functional areas of the building is monitored in real time by installing illuminance sensors at measuring points corresponding to the height of office desks to collect indoor illuminance Lz(t); and the air quality status of the building is monitored in real time by installing carbon dioxide sensors in office areas, meeting rooms, and densely populated areas to collect carbon dioxide concentration Cz(t). The source-side and grid-connected electrical acquisition unit is used to monitor the power generation status of the building photovoltaic (PV) power generation branch in real time. A PV grid-connected detection device is installed on the AC output side of the PV inverter to collect the real-time output power Ppv(t) of the PV. The unit also monitors the power supply and distribution status of the connection point between the building and the external power grid in real time. A power quality detection device is installed in the distribution cabinet at the grid connection point to collect the grid connection point voltage Ug(t) and the active power Pg(t). Based on the historical PV output data, a PV output time series prediction method is used to obtain the predicted PV output power Ppvf(t). The rated voltage UN at the grid connection point is obtained from the building grid connection agreement file. The unit also receives the grid demand response reduction power Pdr(t) issued by the grid dispatching platform by installing a grid demand response terminal in the building energy management room. The load acquisition and reduction calculation unit is used to monitor the overall building power load in real time. A smart energy meter is installed in the main incoming power distribution cabinet to collect the total building power consumption PL(t). The target building load power Ptar(t) is obtained from load control commands issued by the power grid demand response terminal, building energy consumption plan data, or peak load control values ​​preset by maintenance personnel. The power consumption status of critical building loads is monitored in real time. A smart energy meter is installed at the outgoing end of the critical load distribution branch to collect the critical load power Pk(t). The power consumption status of flexible building loads is monitored in real time. Sub-meters are installed in the air conditioning distribution branch, fresh air distribution branch, lighting distribution branch, and office equipment distribution branch to collect the flexible load power Pf(t). Based on the flexible load power Pf(t), the number of people Nz(t), indoor temperature Tz(t), indoor illuminance Lz(t), and carbon dioxide concentration Cz(t), a comfort constraint reduction calculation method is used to obtain the flexible load reductionable power Pcut(t). The energy storage status and power acquisition unit is used to monitor the battery operating status of the building energy storage equipment in real time. It connects to the battery manager in the building energy storage battery cluster control cabinet to acquire the energy storage state of charge (SOC(t), energy storage health status (SOH(t)) and energy storage rated capacity (EN). It also monitors the charging and discharging power status of the building energy storage equipment in real time. A bidirectional energy meter is installed on the AC output side of the energy storage converter to acquire the current charging power Pch(t), the current discharging power Pdis(t), and the maximum allowable discharging power Pdis,max(t). The power distribution branch load acquisition unit is used to monitor the load status of each power distribution branch participating in source load matching in real time. Current transformers are installed at the outgoing cables of each power distribution branch to collect the real-time current of each power distribution branch. Obtain the rated current of each power distribution branch from the building's power distribution design documents. .

[0026] In this embodiment, by collecting data on the number of people, indoor temperature, indoor illuminance, carbon dioxide concentration, photovoltaic output power, grid connection point electrical data, building load data, energy storage operation data, and power distribution branch load data, the system can obtain complete basic operational data from the source side, load side, energy storage side, power distribution side, and comfort side, thus avoiding the problem of incomplete control basis caused by existing building control systems relying only on local energy consumption data.

[0027] Example 3 Please see Figure 1 In the explanation of Embodiment 2, the source-load topology matching evaluation module specifically includes a source-load power balance calculation unit, a distribution branch load calculation unit, an energy storage support capacity calculation unit, a flexible load adjustment calculation unit, a source-load topology matching index calculation unit, and a first analysis unit. The source-load power balance calculation unit is used to extract the total building power consumption PL(t), the real-time output power of building photovoltaics Ppv(t), the maximum allowable discharge power of energy storage Pdis,max(t), and the grid demand response reduction power Pdr(t), and calculate the source-load power balance coefficient Cb(t). The calculation formula is as follows:

[0028] In the formula, This represents a very small constant to prevent the denominator from being zero; The physical principle of the formula is to reflect the degree of power balance between the current building power supply capacity and load demand by comparing the difference between the total power consumption of the building, the real-time output power of the building photovoltaic system, the maximum allowable discharge power of the energy storage, and the power reduction power of the grid demand response.

[0029] The power distribution branch load calculation unit is used to extract the real-time current of the i-th power distribution branch. and the rated current of the i-th distribution branch The remaining load factor Br(t) of the distribution branch is calculated using the following formula:

[0030] In the formula, n represents the number of distribution branches participating in source-load matching; The physical principle of the formula is to reflect the safe ability of the building's power distribution branches to continue to carry photovoltaic, energy storage, or load regulation power during the source-load matching process by comparing the residual margin between the real-time current of each distribution branch and the corresponding rated current.

[0031] The energy storage support capacity calculation unit is used to extract the energy storage state of charge (SOC(t)) and energy storage health state (SOH(t)). An interval normalization method is used to normalize the SOC(t) and SOH(t) to obtain the normalized SOCn(t) and SOHn(t) respectively. Combined with the rated energy storage capacity EN, the available energy storage capacity Es(t) is calculated using the following formula:

[0032] The physical principle of the formula is to reflect the effective capacity of the energy storage system that can actually be used to support the operation of the building under the current power level and battery health conditions by combining the rated capacity of the energy storage with the normalized state of charge and the normalized state of health of the energy storage.

[0033] The energy storage support coefficient Sr(t) is calculated using the available energy storage capacity Es(t) and the rated energy storage capacity EN, as shown in the following formula:

[0034] The physical principle of the formula is to reflect the current ability of the energy storage system to support peak shaving, power supply and source-load balance of buildings by comparing the ratio between the available energy storage capacity and the rated energy storage capacity.

[0035] The flexible load adjustment calculation unit extracts the flexible load reduction power Pcut(t) and the flexible load power Pf(t), and calculates the flexible load adjustment coefficient Fr(t) using the following formula:

[0036] In the formula, This represents a minimal constant to prevent the denominator from being zero.

[0037] The physical principle of the formula is to reflect the ability of flexible loads such as air conditioning, fresh air, lighting and office equipment to participate in reduction and adjustment under comfort constraints by comparing the proportional relationship between the power that can be reduced by the flexible load and the power of the flexible load.

[0038] In this embodiment, by calculating the source-load power balance coefficient, the remaining load capacity coefficient of the distribution branch, the energy storage support coefficient, and the flexible load adjustment coefficient, the current power supply capacity, distribution load capacity, energy storage support capacity, and flexible load adjustment capacity of the building can be quantitatively expressed, providing a calculable and comparable basis for subsequent judgment on whether the building source-load topology matches.

[0039] Example 4 Please see Figure 1 In the explanation of Embodiment 3, specifically, the source-load topology matching index calculation unit is used to calculate the source-load topology matching index Ms(t) after dimensionless processing of the obtained source-load power balance coefficient Cb(t), distribution branch residual load factor Br(t), energy storage support factor Sr(t), and flexible load adjustment factor Fr(t). The calculation formula is as follows:

[0040] In the formula, w1, w2, w3, and w4 represent weighting coefficients; The source-load power balance coefficient Cb(t) represents the impact of the source-load topology matching index Ms(t) and has the highest weight. The source-load power balance coefficient directly reflects the supply-demand matching relationship between the total building power consumption, the real-time output power of building photovoltaics, the maximum allowable discharge power of energy storage, and the power reduction power of grid demand response. It is the core basis for judging whether the building source-load topology is in a stable operating state. When the source-load power balance capacity is insufficient, even if the distribution branches still have carrying capacity, there may be insufficient power supply to critical loads or difficulties in demand response execution. Therefore, it is given a high weight. The residual carrying capacity factor Br(t) of the distribution branch characterizes the impact of the source-load topology matching index Ms(t), and has a medium to high weight. The residual carrying capacity factor of the distribution branch reflects whether each distribution branch participating in source-load matching still has a current carrying margin, which directly affects whether safe power supply path adjustments can be made between photovoltaic branches, energy storage branches, and load branches. When the distribution branch is close to its rated carrying capacity, even if the source-side power is sufficient, it is not advisable to continue to increase the power supply burden of the branch. Therefore, this parameter is used to constrain the distribution safety boundary in the source-load matching process. The energy storage support coefficient Sr(t) characterizes the impact of the source-load topology matching index Ms(t), and has a medium-to-high weight. The energy storage support coefficient reflects the effective support capacity that the energy storage system can provide under the current state of charge and health of the energy storage system. It is an important adjustment resource for buildings to cope with photovoltaic fluctuations, peak loads, and demand response reduction requirements. When the energy storage support capacity is high, the system has stronger peak shaving, supply guarantee, and source-load balancing capabilities, so its weight is higher than that of the flexible load adjustment coefficient. The flexible load adjustment coefficient Fr(t) characterizes the impact of the source-load topology matching index Ms(t), and has a relatively low weight. The flexible load adjustment coefficient reflects the ability of flexible loads such as air conditioning, fresh air, lighting, and office equipment to be reduced without compromising indoor comfort constraints. Since flexible load adjustment is limited by the number of people, indoor temperature, indoor illuminance, and carbon dioxide concentration, it cannot be reduced indefinitely. Therefore, its role is more reflected in auxiliary balancing and local regulation, and its weight is lower than that of parameters related to source-load power balance and energy storage support. By constructing a source-load topology matching index Ms(t), which is a weighted fusion of source-load supply and demand balance, distribution capacity margin, energy storage support capacity, and flexible load adjustability, the power supply matching relationship among building photovoltaic branches, energy storage branches, critical load branches, and flexible load branches can be comprehensively reflected. When the value of Ms(t) increases, it indicates that the matching between the building's source-side power supply capacity, energy storage support capacity, and load-side adjustment capacity is good, and the building's source-load topology tends to be stable. When Ms(t) is low, it indicates that the system may have problems such as source-load supply and demand imbalance, insufficient distribution branch capacity, insufficient energy storage support capacity, or limited flexible load adjustment space, which requires triggering a local distribution reconfiguration strategy or a forced source-load matching strategy.

[0041] The first analysis unit is used to obtain a first evaluation result by comparing the source-load topology matching index Ms(t) with the first matching threshold M1 and the second matching threshold M2 through preset first matching threshold M1 and second matching threshold M2: When the source-load topology matching index Ms(t) ≥ the first matching threshold M1, the building source-load topology is determined to be in a matching state. The existing photovoltaic power supply path, energy storage operation status and flexible load operation status are maintained and continuously monitored. When the second matching threshold M2 ≤ source-load topology matching index Ms(t) < the first matching threshold M1, the building source-load topology is determined to be in a first-level mismatch state, and a local power distribution reconfiguration strategy is generated: adjust the power supply matching relationship between photovoltaic branches, energy storage branches and flexible load branches, so that the energy storage system prioritizes supporting critical load branches and peak reduction branches, and performs amplitude limiting adjustment on flexible loads; When the source-load topology matching index Ms(t) < the second matching threshold M2, the building source-load topology is determined to be in a level 2 mismatch state, and a forced source-load matching strategy is generated: control the energy storage system to enter the discharge support state, reduce the power of non-critical loads, and maintain the power supply priority of critical loads.

[0042] The method for obtaining the first matching threshold M1 and the second matching threshold M2: The goal of the calibration process of the first matching threshold M1 and the second matching threshold M2 is to determine two critical source load topology matching index Ms(t) values ​​that can effectively distinguish between "building source load topology matching state", "first-level mismatch state" and "second-level mismatch state".

[0043] A database of standard data for building energy source-load topology matching is constructed. This database is formed through the collection, simulation, and statistical analysis of building energy operation data under different seasons, workday types, photovoltaic output levels, building load levels, energy storage operation states, and distribution branch load states. The database has a sample size of no less than 1000 sets, covering various operating conditions such as energy source-load supply and demand balance, mild energy source-load mismatch, and severe energy source-load mismatch. Each sample in the database includes the building's total power consumption, real-time output power of building photovoltaics, maximum allowable discharge power of energy storage, power reduction power of grid demand response, real-time current of each distribution branch, rated current of each distribution branch, energy storage state of charge, energy storage health status, energy storage rated capacity, power reduction power of flexible load, and power of flexible load.

[0044] Meanwhile, three or more technical personnel with experience in building energy management, power distribution system operation, or energy storage dispatch will use a "gold standard" to label the source-load topology status of each sample based on the photovoltaic branch power supply capacity, energy storage support capacity, critical load power supply status, remaining carrying capacity of the power distribution branch, and the adjustability of flexible loads. This will clearly classify the status as "building source-load topology matching status," "Level 1 mismatch status," or "Level 2 mismatch status." A building source-load topology matching status indicates that the existing photovoltaic power supply path, energy storage operation status, and flexible load operation status can meet the current building energy demand; a Level 1 mismatch status indicates the need for partial power distribution reconfiguration and flexible load limiting adjustment; and a Level 2 mismatch status indicates the need for forced source-load matching and reduction of non-critical load power.

[0045] All sample data in the database were uniformly calculated using the same method as the source-load topology matching index Ms(t) to obtain the source-load topology matching index value for each sample group. Next, statistical distribution analysis was performed on the source-load topology matching index values ​​for three types of samples: "building source-load topology matching state," "first-level mismatch state," and "second-level mismatch state," and probability density function curves were plotted. Theoretically, the source-load topology matching index value for the building source-load topology matching state is generally higher, the source-load topology matching index value for the first-level mismatch state is in the middle range, and the source-load topology matching index value for the second-level mismatch state is generally lower.

[0046] The classification performance was evaluated by constructing ROC curves. "Building source-load topology matching state" and "mismatch state" were used as binary classification objects, and the point with the largest Youden index was selected as the first matching threshold M1. Then, "Level 1 mismatch state" and "Level 2 mismatch state" were used as binary classification objects, and the point with the largest Youden index was selected as the second matching threshold M2, thus achieving a balance between sensitivity and specificity in the graded determination of source-load topology matching state.

[0047] Based on the above statistical analysis and operational experience, the preferred first matching threshold M1 is 0.75, and the preferred second matching threshold M2 is 0.45. When the source-load topology matching index Ms(t) is not lower than 0.75, the building source-load topology is determined to be in a matched state, and the current topology maintenance strategy is executed; when the source-load topology matching index Ms(t) is not lower than 0.45 but lower than 0.75, the building source-load topology is determined to be in a first-level mismatch state, and a local power distribution reconfiguration strategy is triggered; when the source-load topology matching index Ms(t) is lower than 0.45, the building source-load topology is determined to be in a second-level mismatch state, and a forced source-load matching strategy is triggered.

[0048] In this embodiment, the source-load topology matching index is calculated by integrating the source-load power balance coefficient, the remaining load factor of the distribution branch, the energy storage support coefficient, and the flexible load adjustment coefficient. Based on the threshold, the current topology maintenance strategy, the local distribution reconfiguration strategy, or the forced source-load matching strategy are generated, so that the system can take different control methods according to different mismatches, thereby improving the supply capacity of critical loads and reducing the power supply mismatch between photovoltaic branches, energy storage branches, and flexible load branches.

[0049] Example 5 Please see Figure 1 In the explanation of Embodiment 4, the energy storage virtual sub-slot scheduling module specifically includes a photovoltaic surplus power calculation unit, a peak reduction and supply guarantee demand calculation unit, a comfort guarantee demand calculation unit, an energy storage virtual sub-slot scheduling coefficient calculation unit, and a second analysis unit. The photovoltaic surplus power calculation unit is used to extract the real-time output power Ppv(t) of the building photovoltaic system and the total power consumption PL(t) of the building. It uses a non-negative truncation photovoltaic surplus calculation method to calculate the photovoltaic surplus power Psur(t). The calculation formula is as follows:

[0050] The physical principle of the formula is to extract the non-negative part of the real-time output power of building photovoltaics that exceeds the total power consumption of the building, so as to reflect the current building photovoltaic power generation capacity after meeting the local load.

[0051] The peak reduction and supply guarantee demand calculation unit is used to extract the grid demand response reduction power Pdr(t), the total building power consumption PL(t), and the building target load power Ptar(t). It then uses the peak difference calculation method to calculate the peak reduction demand power Ppeak(t), as shown in the following formula:

[0052] The physical principle of the formula is to reflect the power demand of the building that needs to be suppressed by energy storage discharge or load reduction by comparing the relationship between the building's total power consumption, the building's target load power, and the power reduction power of the grid demand response.

[0053] The comfort guarantee requirement calculation unit is used to extract indoor temperature Tz(t), indoor illuminance Lz(t), carbon dioxide concentration Cz(t), number of people Nz(t), and flexible load power Pf(t). It employs a comfort boundary deviation calculation method to calculate the degree of deviation of indoor temperature, indoor illuminance, carbon dioxide concentration, and number of people from the preset comfort boundary, obtaining the comfort deviation coefficient Dcom(t). Combined with the flexible load power Pf(t), the comfort guarantee requirement power Pcom(t) is calculated using the following formula:

[0054] The physical principle of the formula is to combine the flexible load power with the comfort deviation coefficient to reflect the flexible load power requirements such as air conditioning, fresh air, and lighting that need to be retained or reserved to maintain the comfort of personnel.

[0055] In this embodiment, by calculating the surplus photovoltaic power, peak reduction demand power, comfort deviation coefficient, and comfort guarantee demand power, energy storage scheduling no longer relies solely on the energy storage state of charge or peak-valley electricity price for charging and discharging. Instead, it can simultaneously respond to photovoltaic consumption, peak shaving demand, critical load supply guarantee, and indoor comfort guarantee demand, thereby improving the targeting of energy storage capacity allocation.

[0056] Example 6 Please see Figure 1 In the explanation of Example 5, specifically, the energy storage virtual sub-slot scheduling coefficient calculation unit is used to calculate the energy storage virtual sub-slot scheduling coefficient Ke(t) by combining the acquired photovoltaic surplus power Psur(t), peak reduction demand power Ppeak(t), and comfort guarantee demand power Pcom(t), along with the critical load power Pk(t) and the total building power consumption PL(t), after dimensionless processing. The calculation formula is as follows:

[0057] In the formula, s1, s2, s3 and s4 represent weighting coefficients; The photovoltaic surplus power Psur(t) characterizes the impact of the energy storage virtual storage compartment dispatch coefficient Ke(t), and has a medium weight. The photovoltaic surplus power reflects the available capacity after the real-time output power of the building photovoltaic exceeds the total power consumption of the building. When the photovoltaic surplus power increases, the energy storage system needs to prioritize the charging demand of the photovoltaic virtual storage compartment to reduce photovoltaic curtailment and backflow pressure. However, photovoltaic surplus is usually a demand for absorbing available energy, and its urgency is slightly lower than that of peak load reduction and critical load supply, so it is given a medium weight. This parameter represents the impact of peak demand reduction power Ppeak(t) on the energy storage virtual compartment scheduling coefficient Ke(t), and has the highest weight. Peak demand reduction power directly reflects the degree to which the current building load exceeds the building target load power or the grid demand response reduction power requirement, and is an important basis for whether the energy storage system needs to immediately enter the discharge support state. When the peak demand reduction is large, if the storage cannot support it in time, it will affect the demand response execution effect and the building peak load control effect, therefore it is given the highest weight. The parameter Pcom(t) represents the impact of the comfort guarantee demand power on the energy storage virtual compartment scheduling coefficient Ke(t), and has a relatively low weight. The comfort guarantee demand power reflects the flexible load power required to maintain indoor comfort in a building under constraints of indoor temperature, indoor illuminance, carbon dioxide concentration, and number of occupants. This parameter is used to avoid excessive reduction of air conditioning, fresh air, and lighting equipment during peak shaving or demand response, but its scheduling priority is usually lower than that of critical load supply and peak shaving, so it is set as an auxiliary guarantee weight. This represents the impact of critical load power Pk(t) on the energy storage virtual compartment scheduling coefficient Ke(t), and has a high weight. Critical load power reflects the load demand within a building that cannot be arbitrarily reduced or interrupted, such as fire protection, security, core office equipment, information rooms, or other important power branches. Ensuring the supply of critical loads is crucial to building operational safety and the continuity of basic functions. When the proportion of critical loads increases, the energy storage system needs to reserve more capacity for the critical load supply virtual compartment, hence its high weight. By constructing a virtual energy storage allocation coefficient Ke(t) that weights and integrates photovoltaic (PV) absorption demand, peak shaving demand, comfort assurance demand, and critical load supply assurance demand, the urgency of capacity allocation for the energy storage system among different operational objectives can be comprehensively reflected. When Ke(t) increases, it indicates that the energy storage system faces strong peak shaving, supply assurance, or allocation demands, and capacity should be prioritized for allocation to the peak shaving virtual storage and critical load supply virtual storage. When Ke(t) is in the middle range, it indicates that the energy storage system can simultaneously accommodate the PV absorption virtual storage and the comfort assurance virtual storage. When Ke(t) is low, it indicates that maintaining the current charge / discharge state of the energy storage system is sufficient to meet the building's operational needs.

[0058] The second analysis unit is used to obtain a second evaluation result by presetting a first sub-storage threshold K1 and a second sub-storage threshold K2, and comparing the energy storage virtual sub-storage scheduling coefficient Ke(t) with the first sub-storage threshold K1 and the second sub-storage threshold K2. When the energy storage virtual sub-compartment scheduling coefficient Ke(t) ≥ the first sub-compartment threshold K1, the energy storage system is determined to be in the first-level scheduling demand state, and the first sub-compartment scheduling strategy is generated: the available energy storage capacity is preferentially allocated to the peak reduction virtual compartment and the critical load supply virtual compartment, the energy storage system is controlled to discharge to the critical load branch and the peak reduction branch, and the energy storage discharge support command is output. When the second sub-storage threshold K2 ≤ energy storage virtual sub-storage scheduling coefficient Ke(t) < the first sub-storage threshold K1, the energy storage system is determined to be in a secondary scheduling demand state, and a second sub-storage scheduling strategy is generated: the available energy storage capacity is allocated to the photovoltaic consumption virtual warehouse and the comfort guarantee virtual warehouse; when the photovoltaic surplus power is greater than zero, the energy storage system is controlled to charge; when the comfort guarantee demand power increases, the corresponding energy storage capacity is reserved, and an energy storage charging or capacity reservation command is output. When the energy storage virtual sub-slot scheduling coefficient Ke(t) < the second sub-slot threshold K2, the energy storage system is determined to be in a level 3 scheduling demand state, and a third sub-slot scheduling strategy is generated: maintain the current energy storage charging and discharging state, maintain the existing energy storage capacity allocation ratio, and continuously monitor the surplus photovoltaic power, peak reduction demand power, comfort guarantee demand power, and critical load power.

[0059] The method for obtaining the first sub-storage threshold K1 and the second sub-storage threshold K2: The goal of the calibration process of the first sub-storage threshold K1 and the second sub-storage threshold K2 is to determine the two critical energy storage virtual sub-storage scheduling coefficients Ke(t) that can effectively distinguish between "Level 1 scheduling demand state", "Level 2 scheduling demand state" and "Level 3 scheduling demand state".

[0060] A database of standard data for virtual energy storage dispatching was constructed. This database was formed through the collection, simulation, and statistical analysis of energy storage dispatching data under different photovoltaic surplus states, different peak reduction demands, different comfort guarantee demands, different critical load supply demands, and different total building electricity consumption levels. The database has a sample size of no less than 800 sets, covering various operating conditions such as energy storage strong discharge support demand, energy storage charging or capacity reservation demand, and energy storage maintenance operation demand. Each sample in the database includes the real-time output power of building photovoltaics, total building electricity consumption, grid demand response reduction power, building target load power, indoor temperature, indoor illuminance, carbon dioxide concentration, number of people, flexible load power, and critical load power.

[0061] Simultaneously, three or more technical personnel with experience in building energy storage dispatch, demand response control, or integrated energy management will use a "gold standard" to label the energy storage dispatch demand status of each sample based on the allocation results of surplus photovoltaic power, peak demand reduction power, comfort guarantee power, critical load power, and available energy storage capacity. This will clearly classify the status into "Level 1 Dispatch Demand Status," "Level 2 Dispatch Demand Status," or "Level 3 Dispatch Demand Status." Level 1 Dispatch Demand Status indicates that the energy storage system should prioritize supporting the peak demand reduction virtual warehouse and the critical load guarantee virtual warehouse; Level 2 Dispatch Demand Status indicates that the energy storage system should be allocated to the photovoltaic consumption virtual warehouse and the comfort guarantee virtual warehouse; Level 3 Dispatch Demand Status indicates that the energy storage system can meet operational needs by maintaining its current charge / discharge state and existing capacity allocation ratio.

[0062] All sample data in the database were uniformly calculated using the same method as the energy storage virtual compartment scheduling coefficient Ke(t) to obtain the energy storage virtual compartment scheduling coefficient value for each sample group. Then, a statistical distribution analysis was performed on the energy storage virtual compartment scheduling coefficient values ​​for the three types of samples: "Level 1 scheduling demand state," "Level 2 scheduling demand state," and "Level 3 scheduling demand state," and probability density function curves were plotted. Theoretically, the energy storage virtual compartment scheduling coefficient value for Level 1 scheduling demand state is generally higher, the value for Level 2 scheduling demand state is in the middle range, and the value for Level 3 scheduling demand state is generally lower.

[0063] The classification performance was evaluated by constructing ROC curves. "Level 1 scheduling demand status" and "non-Level 1 scheduling demand status" were used as the binary classification objects, and the point with the largest Youden index was selected as the first sub-classification threshold K1. Then, "Level 2 scheduling demand status" and "Level 3 scheduling demand status" were used as the binary classification objects, and the point with the largest Youden index was selected as the second sub-classification threshold K2, thus achieving a balance between sensitivity and specificity in the classification of energy storage sub-classification scheduling demand.

[0064] Based on the above statistical analysis and operational experience, the optimal threshold values ​​for the first sub-storage (K1) are 0.70 and for the second sub-storage (K2) are 0.35. When the virtual sub-storage scheduling coefficient Ke(t) is not lower than 0.70, the energy storage system is determined to be in a first-level scheduling demand state, and a first sub-storage scheduling strategy is generated. When the virtual sub-storage scheduling coefficient Ke(t) is not lower than 0.35 but lower than 0.70, the energy storage system is determined to be in a second-level scheduling demand state, and a second sub-storage scheduling strategy is generated. When the virtual sub-storage scheduling coefficient Ke(t) is lower than 0.35, the energy storage system is determined to be in a third-level scheduling demand state, and a third sub-storage scheduling strategy is generated.

[0065] In this embodiment, by calculating the energy storage virtual compartment scheduling coefficient and making a graded judgment, the available energy storage capacity can be dynamically allocated among the peak reduction virtual compartment, the critical load supply virtual compartment, the photovoltaic consumption virtual compartment, and the comfort guarantee virtual compartment, thereby alleviating the problems of insufficient energy storage utilization when photovoltaics are in surplus, untimely energy storage support during peak loads, and conflicts between comfort guarantee and demand response.

[0066] Example 7 Please see Figure 1 In the explanation of Embodiment Six, the source load fluctuation suppression assessment module specifically includes a grid connection point power fluctuation calculation unit, a grid connection point voltage offset calculation unit, a photovoltaic output change calculation unit, a building load change calculation unit, a source load fluctuation suppression coefficient calculation unit, and a third analysis unit. The grid connection point power fluctuation calculation unit is used to extract the grid connection point active power Pg(t) at the current sampling time and the grid connection point active power Pg(t) at the previous sampling time. Δt) and the total building power consumption PL(t) are used to calculate the power fluctuation coefficient Gp(t) at the grid connection point using the sampling period difference calculation method. The calculation formula is as follows:

[0067] In the formula, Indicates the preset sampling period. This represents a very small constant to prevent the denominator from being zero; The physical principle of the formula is to compare the change in active power at the grid connection point between the current sampling time and the previous sampling time, and normalize it by combining it with the total power consumption of the building, so as to reflect the degree of fluctuation in power exchange between the building and the external power grid.

[0068] The grid connection point voltage offset calculation unit is used to extract the grid connection point voltage Ug(t) and the grid connection point rated voltage UN at the current sampling time, and calculates the grid connection point voltage offset coefficient Gu(t) using the rated voltage deviation ratio calculation method. The calculation formula is as follows:

[0069] The physical principle of the formula is to reflect the degree of deviation of the building's grid connection point voltage from the normal operating voltage by comparing the deviation ratio between the grid connection point voltage and the grid connection point rated voltage.

[0070] The photovoltaic output change calculation unit is used to extract the real-time output power Ppv(t) of the building photovoltaic system at the current sampling time and the output power Ppv(t) of the building photovoltaic system at the previous sampling time. The photovoltaic output variation coefficient Gpv(t) is calculated using the photovoltaic output differential change rate calculation method. The calculation formula is as follows:

[0071] In the formula, Indicates the preset sampling period. This represents a very small constant to prevent the denominator from being zero; The physical principle of the formula is to compare the change in building photovoltaic output power between the current sampling time and the previous sampling time, and to normalize it by combining the photovoltaic output power at the previous sampling time, so as to reflect the fluctuation intensity of the photovoltaic power generation side in adjacent sampling periods.

[0072] The building load change calculation unit is used to extract the total building power consumption PL(t) at the current sampling time and the total building power consumption PL(t) at the previous sampling time. The building load variation coefficient GL(t) is calculated using the differential rate of change method for building load (Δt). The calculation formula is as follows:

[0073] In the formula, Indicates the preset sampling period. This represents a minimal constant to prevent the denominator from being zero.

[0074] The physical principle of the formula is to compare the change in the total building power consumption at the current sampling time with that at the previous sampling time, and to normalize it by combining the total building power consumption at the previous sampling time, so as to reflect the intensity of the change in the building load side within adjacent sampling periods.

[0075] In this embodiment, by calculating the grid connection point power fluctuation coefficient, grid connection point voltage deviation coefficient, photovoltaic output change coefficient, and building load change coefficient respectively, the system can identify the source of building load fluctuation from three aspects: grid connection point status, photovoltaic side changes, and load side changes, providing a basis for fluctuation identification for energy storage short-term compensation, voltage support, and flexible load gradual change control.

[0076] Example 8 Please see Figure 1 In the explanation of Example 7, specifically, the source-load fluctuation suppression coefficient calculation unit is used to calculate the source-load fluctuation suppression coefficient Ws(t) after dimensionless processing of the obtained grid connection point power fluctuation coefficient Gp(t), grid connection point voltage deviation coefficient Gu(t), photovoltaic output variation coefficient Gpv(t), and building load variation coefficient GL(t). The calculation formula is as follows:

[0077] In the formula, r1, r2, r3 and r4 represent weighting coefficients.

[0078] The grid-connected power fluctuation coefficient Gp(t) represents the impact of the source-load fluctuation suppression coefficient Ws(t) and has a high weight. The grid-connected power fluctuation coefficient directly reflects the degree of change in active power exchange between the building and the external power grid, and is a key indicator for judging whether source-load fluctuations have been transmitted to the grid-connected point. When the active power fluctuation at the grid-connected point is large, it indicates that the power balance between photovoltaic, energy storage and load inside the building is disturbed, and short-term power compensation by the energy storage converter is required. Therefore, it is given a high weight. The grid connection point voltage deviation coefficient Gu(t) characterizes the impact of the source load fluctuation suppression coefficient Ws(t) and has a high weight. The grid connection point voltage deviation coefficient reflects the degree of deviation of the building's grid connection point voltage from the rated voltage, which is directly related to the operational safety of the power distribution system and the stability of electrical equipment. When the grid connection point voltage deviation increases, it indicates that power fluctuations inside the building may cause voltage support demand, requiring the energy storage converter to enter voltage support mode. Therefore, this parameter has a high control priority. The photovoltaic output variation coefficient Gpv(t) characterizes the impact of the source load fluctuation suppression coefficient Ws(t), and has a medium weight. The photovoltaic output variation coefficient reflects the degree of change in the real-time output power of building-integrated photovoltaics within adjacent sampling periods and is the main source of source-side fluctuations. Since photovoltaic output fluctuations need to be judged in conjunction with changes in building load and grid connection point response, photovoltaic changes alone do not necessarily directly cause grid connection point disturbances. Therefore, its weight is lower than that of the grid connection point power fluctuation coefficient and the grid connection point voltage deviation coefficient. The building load variation coefficient GL(t) represents the impact of the source load fluctuation suppression coefficient Ws(t) and has a relatively low weight. The building load variation coefficient reflects the degree of change in the total building power consumption within adjacent sampling periods, mainly reflecting the load-side disturbances caused by the start-up and shutdown of air conditioning, fresh air, lighting, and office equipment. Since some load changes can be smoothed through flexible load variation strategies, amplitude limiting strategies, and short-term energy storage compensation strategies, its weight is lower than that of the direct electrical state-related parameters at the grid connection point. By constructing a source-load fluctuation suppression coefficient Ws(t) that is a weighted fusion of grid-connected power fluctuations, grid-connected voltage deviations, photovoltaic output changes, and building load changes, the impact of building source-load fluctuations on grid-connected stability can be comprehensively reflected. When the value of Ws(t) increases, it indicates that the fluctuations on the building source side or load side have had a significant impact on the grid-connected power or voltage, requiring the triggering of short-term energy storage compensation strategies or voltage support and load mitigation strategies. When Ws(t) is low, it indicates that the current building source-load fluctuations are within the allowable range, and the current energy storage charging and discharging state, photovoltaic power supply path, and flexible load operation state can be maintained.

[0079] In this embodiment, the source load fluctuation suppression coefficient is formed by combining the grid connection point power fluctuation coefficient, grid connection point voltage deviation coefficient, photovoltaic output change coefficient and building load change coefficient. This enables the system to uniformly evaluate the degree of building source load fluctuation, avoiding misjudgment based solely on a single voltage or power change, thereby improving the accuracy and stability of grid connection point fluctuation suppression judgment.

[0080] Example 9 Please see Figure 1 In the explanation of Embodiment Eight, specifically, the third analysis unit is used to compare and analyze the source load fluctuation suppression coefficient Ws(t) with the first fluctuation threshold W1 and the second fluctuation threshold W2 by setting a first fluctuation threshold W1 and a second fluctuation threshold W2, and obtain a third evaluation result, including: When the source load fluctuation suppression coefficient Ws(t) < the first fluctuation threshold W1, the building source load fluctuation is determined to be in an allowable state, and a fluctuation allowance strategy is generated: maintain the current energy storage charging and discharging state, photovoltaic power supply path and flexible load operation state, and continuously monitor; When the first fluctuation threshold W1 ≤ source load fluctuation suppression coefficient Ws(t) < the second fluctuation threshold W2, the building source load fluctuation is determined to be in the first-level suppression state, and a short-term energy storage compensation strategy is generated: control the energy storage converter to output short-term compensation power within the range allowed by the current energy storage compartment scheduling state, suppress the active power fluctuation of the grid connection point, and limit the power adjustment range of flexible loads. When the source load fluctuation suppression coefficient Ws(t) ≥ the second fluctuation threshold W2, the building source load fluctuation is determined to be in the second-level suppression state, and a voltage support and load slow change strategy is generated: control the energy storage converter to enter the voltage support state, limit the power rise rate of air conditioning, fresh air and lighting equipment, reduce the start-up priority of non-critical loads, and maintain the power supply stability of critical load branches.

[0081] The method for obtaining the first fluctuation threshold W1 and the second fluctuation threshold W2: The goal of the calibration process of the first fluctuation threshold W1 and the second fluctuation threshold W2 is to determine the two critical source load fluctuation suppression coefficients Ws(t) values ​​that can effectively distinguish between the "permissible state of building source load fluctuation", "first-level suppression state" and "second-level suppression state".

[0082] A database of standard data for suppressing building load fluctuations was constructed. This database was formed through the collection, simulation, and statistical analysis of building grid-connected operation data under different photovoltaic output fluctuations, different building load abrupt changes, different grid-connected point active power changes, different grid-connected point voltage deviations, and different energy storage converter response states. The database has a sample size of no less than 1000 sets, covering various operating conditions such as fluctuation-allowed states, short-term power compensation demand states, and voltage support and load gradual change demand states. Each sample in the database includes the grid-connected point active power at the current sampling time, the grid-connected point active power at the previous sampling time, the grid-connected point voltage, the grid-connected point rated voltage, the real-time output power of the building's photovoltaic system at the current sampling time, the output power of the building's photovoltaic system at the previous sampling time, the total building power consumption at the current sampling time, and the total building power consumption at the previous sampling time.

[0083] Meanwhile, three or more technical personnel with experience in building power distribution operation, power quality analysis, or energy storage converter control will use a "gold standard" to label the source-load fluctuation state of each sample based on the grid connection point power fluctuation amplitude, grid connection point voltage deviation, photovoltaic output variation, building load variation, and energy storage short-term compensation effect. This will clearly classify the state into "permissible building source-load fluctuation state," "Level 1 suppression state," or "Level 2 suppression state." The permissible building source-load fluctuation state indicates that there is no need to change the current energy storage charging / discharging state, photovoltaic power supply path, or flexible load operation state. The Level 1 suppression state indicates that short-term energy storage compensation is needed to suppress active power fluctuations at the grid connection point. The Level 2 suppression state indicates that the energy storage converter needs to be controlled to enter a voltage support state, and the rate of increase of flexible load power needs to be limited.

[0084] All sample data in the database were uniformly calculated using the same method as the source load fluctuation suppression coefficient Ws(t) to obtain the source load fluctuation suppression coefficient value for each sample group. Next, a statistical distribution analysis was performed on the source load fluctuation suppression coefficient values ​​for the three types of samples: "allowable source load fluctuation state," "Level 1 suppression state," and "Level 2 suppression state," and probability density function curves were plotted. Theoretically, the source load fluctuation suppression coefficient value for the "allowable source load fluctuation state" is generally lower, the value for the "Level 1 suppression state" is in the middle range, and the value for the "Level 2 suppression state" is generally higher.

[0085] The classification performance was evaluated by constructing ROC curves. "Allowable state of building load fluctuation" and "Suppressed state" were used as binary classification objects, and the point with the largest Youden index was selected as the first fluctuation threshold W1. Then, "Level 1 Suppressed State" and "Level 2 Suppressed State" were used as binary classification objects, and the point with the largest Youden index was selected as the second fluctuation threshold W2, thus achieving a balance between sensitivity and specificity in the classification of building load fluctuation suppression state.

[0086] Based on the above statistical analysis and operational experience, the preferred first fluctuation threshold W1 is 0.30, and the preferred second fluctuation threshold W2 is 0.60. When the source load fluctuation suppression coefficient Ws(t) is below 0.30, the building source load fluctuation is determined to be in an allowable state, and a fluctuation allowance strategy is generated; when the source load fluctuation suppression coefficient Ws(t) is not lower than 0.30 and is lower than 0.60, the building source load fluctuation is determined to be in a first-level suppression state, and a short-term energy storage compensation strategy is generated; when the source load fluctuation suppression coefficient Ws(t) is not lower than 0.60, the building source load fluctuation is determined to be in a second-level suppression state, and a voltage support and load gradual change strategy is generated.

[0087] In this embodiment, by comparing the source-load fluctuation suppression coefficient with the first fluctuation threshold and the second fluctuation threshold in a graded manner, and generating fluctuation allowance strategy, energy storage short-term compensation strategy or voltage support and load gradual change strategy, the system can select corresponding control measures according to the severity of fluctuations, thereby reducing the impact of sudden changes in photovoltaic output and sudden increases in building load on grid connection power and voltage stability.

[0088] Example 10 Please see Figure 1 In the explanation of Embodiment Nine, the flexible control strategy generation module specifically includes a comprehensive strategy determination unit and a final control unit. The comprehensive strategy determination unit is used to generate a final execution strategy based on the first evaluation result, the second evaluation result, and the third evaluation result, specifically including: When the third evaluation result is a level 2 suppression state, the voltage support and load gradual change strategy is set as the priority execution strategy; when the first evaluation result is a level 2 mismatch state, the forced source-load matching strategy is set as the priority execution strategy; when the second evaluation result is a level 1 scheduling demand state, the first compartmentalized scheduling strategy is set as the priority execution strategy; when the third evaluation result is an allowable state, the first evaluation result is a matching state, and the second evaluation result is a level 3 scheduling demand state, the current operation maintenance strategy is generated. Conflict handling is performed for energy storage charging commands, energy storage discharging commands, energy storage converter voltage support commands, flexible load limiting commands, and flexible load gradual change commands; when energy storage charging commands and energy storage discharging commands coexist, the energy storage discharging command is executed first; when energy storage discharging support commands and energy storage converter voltage support commands coexist, the capacity required for energy storage converter voltage support is reserved first; when flexible load limiting commands and comfort guarantee capacity reservation commands coexist, the flexible load limiting command is executed. The final control unit is used to generate final building flexible control instructions based on the judgment results, including photovoltaic power supply path control instructions, energy storage compartment capacity call instructions, energy storage charging and discharging control instructions, energy storage converter voltage support instructions, critical load supply guarantee instructions, flexible load limiting instructions, flexible load gradual change instructions, and non-critical load load reduction instructions. It connects to the building power distribution switchgear, energy storage converter, air conditioning controller, fresh air controller, lighting controller and sub-item power equipment controller respectively, and executes the final building flexible control command.

[0089] In this embodiment, the final execution strategy is generated by integrating the first evaluation result, the second evaluation result, and the third evaluation result. Conflict handling is performed on the energy storage charging command, energy storage discharging command, energy storage converter voltage support command, flexible load limiting command, and flexible load gradual change command. This ensures that the final building flexible control command can take into account source-load matching, energy storage compartmentation, fluctuation suppression, and comfort assurance, avoiding command conflicts when multiple control targets are triggered simultaneously.

[0090] It should be noted that all calculation formulas in this application employ regression analysis, including but not limited to machine learning algorithms, to deeply analyze the collected parameters and identify their natural trends and interrelationships. Specialized software, such as Python's Scikit-learn library or the R language, is used to automatically generate mathematical models that match the data. Then, cross-validation and other methods are used to objectively evaluate the model performance, and continuous feedback and optimization are combined to ensure that the created formulas truly reflect the inherent laws of the data, thereby guaranteeing their effectiveness and accuracy. In all calculation formulas in this application, the parameters in each formula undergo dimensionless processing within a consistent range to ensure that different physical quantities are compared on the same scale; dimensionless processing techniques include, but are not limited to, min-max-normalization and Z-score standardization. The algorithm of this invention is implemented as a Python script. Before executing the core logic, the program first executes a data loading module (e.g., using the widely used pandas library in Python) configured to read the aforementioned spreadsheet file and load its contents into the program's working memory (e.g., a DataFrame data structure). Subsequent algorithm steps will directly query and retrieve the required configuration parameters from this in-memory data structure.

[0091] It should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention and are not intended to limit it. Although the present invention has been described in detail with reference to preferred embodiments, those skilled in the art should understand that modifications or equivalent substitutions can be made to the technical solutions of the present invention without departing from the spirit and scope of the technical solutions of the present invention, and all such modifications or substitutions should be covered within the scope of the claims of the present invention.

Claims

1. A flexible intelligent control system for dynamic matching of source and load in buildings, characterized in that, Specifically, it includes: The integrated data acquisition module is used to collect building indoor comfort constraint data, source-side and grid-connected electrical data, load-side load data, energy storage status and power data, and distribution branch load data; The source-load topology matching assessment module is used to calculate the source-load power balance coefficient, the remaining load factor of the distribution branch, the energy storage support coefficient and the flexible load adjustment coefficient, and comprehensively calculate the source-load topology matching index to determine the source-load topology matching status of the building, generate the first assessment result, and output the current topology maintenance strategy, the local power distribution reconfiguration strategy or the forced source-load matching strategy. The energy storage virtual compartment scheduling module is used to calculate the surplus photovoltaic power, peak reduction demand power, comfort deviation coefficient and comfort guarantee demand power, and comprehensively calculate the energy storage virtual compartment scheduling coefficient, determine the compartment scheduling demand status of the energy storage system, generate the second evaluation result, and output the first compartment scheduling strategy, the second compartment scheduling strategy or the third compartment scheduling strategy. The source load fluctuation suppression assessment module is used to calculate the grid connection point power fluctuation coefficient, grid connection point voltage deviation coefficient, photovoltaic output change coefficient and building load change coefficient, and comprehensively calculate the source load fluctuation suppression coefficient, determine the building source load fluctuation suppression status, generate the third assessment result, and output the fluctuation allowance strategy, energy storage short-term compensation strategy or voltage support and load gradual change strategy. The flexible control strategy generation module receives the first assessment result, the second assessment result, and the third assessment result, performs comprehensive strategy determination and control command conflict handling, generates the final execution strategy, and outputs the final building flexible control command.

2. The building flexible intelligent control system for dynamic matching of source and load according to claim 1, characterized in that: The integrated data acquisition module includes an indoor comfort constraint acquisition unit, a source-side and grid-connected electrical acquisition unit, a load-side load acquisition and reduction calculation unit, an energy storage status and power acquisition unit, and a distribution branch load acquisition unit. The indoor comfort constraint acquisition unit is used to collect data on the number of people, indoor temperature, indoor illuminance, and carbon dioxide concentration. The source-side and grid-connected electrical acquisition unit is used to collect the real-time output power of building photovoltaics, grid connection point voltage, grid connection point active power, photovoltaic predicted output power, grid connection point rated voltage, and grid demand response reduction power. The load-side load acquisition and reduction calculation unit is used to acquire the building's total power consumption, target load power, critical load power, and flexible load power, and to calculate the power that can be reduced from the flexible load. The energy storage status and power acquisition unit is used to acquire the energy storage state of charge, energy storage health status, energy storage rated capacity, energy storage current charging power, energy storage current discharging power and energy storage maximum allowable discharge power. The power distribution branch carrying acquisition unit is used to collect the real-time current of each power distribution branch and obtain the rated current of each power distribution branch.

3. The building flexible intelligent control system for dynamic matching of source and load according to claim 2, characterized in that: The source-load topology matching evaluation module includes a source-load power balance calculation unit, a distribution branch load calculation unit, an energy storage support capacity calculation unit, a flexible load adjustment calculation unit, a source-load topology matching index calculation unit, and a first analysis unit; The source-load power balance calculation unit is used to calculate the source-load power balance coefficient based on the building's total power consumption, the real-time output power of the building photovoltaic system, the maximum allowable discharge power of the energy storage system, and the power reduction power of the grid demand response. The power distribution branch load-bearing calculation unit is used to calculate the remaining load-bearing coefficient of each power distribution branch based on the real-time current and rated current of each power distribution branch. The energy storage support capacity calculation unit is used to calculate the available energy storage capacity and energy storage support coefficient based on the energy storage state of charge, energy storage health status and energy storage rated capacity. The flexible load adjustment calculation unit is used to calculate the flexible load adjustment coefficient based on the power reduction capacity of the flexible load and the power of the flexible load.

4. The building flexible intelligent control system for dynamic matching of source and load according to claim 3, characterized in that: The source-load topology matching index calculation unit is used to calculate the source-load topology matching index based on the source-load power balance coefficient, the remaining carrying capacity coefficient of the distribution branch, the energy storage support coefficient, and the flexible load adjustment coefficient. The first analysis unit is used to compare and analyze the source-load topology matching index with a preset first matching threshold and a second matching threshold to obtain a first evaluation result; when the source-load topology matching index reaches the first matching threshold, it is determined that the building source-load topology is in a matching state and a current topology maintenance strategy is generated; when the source-load topology matching index is between the second matching threshold and the first matching threshold, it is determined that the building source-load topology is in a first-level mismatch state and a local power distribution reconfiguration strategy is generated; when the source-load topology matching index is lower than the second matching threshold, it is determined that the building source-load topology is in a second-level mismatch state and a forced source-load matching strategy is generated.

5. The building flexible intelligent control system for dynamic matching of source and load according to claim 4, characterized in that: The energy storage virtual sub-slot scheduling module includes a photovoltaic surplus power calculation unit, a peak reduction and supply guarantee demand calculation unit, a comfort guarantee demand calculation unit, an energy storage virtual sub-slot scheduling coefficient calculation unit, and a second analysis unit. The photovoltaic surplus power calculation unit is used to calculate the photovoltaic surplus power based on the real-time output power of the building photovoltaic system and the total power consumption of the building. The peak reduction and supply guarantee demand calculation unit is used to calculate the peak reduction demand power based on the power reduction power of the power grid demand response, the total power consumption of the building, and the target load power of the building; The comfort guarantee requirement calculation unit is used to calculate the comfort deviation coefficient and comfort guarantee requirement power based on indoor temperature, indoor illuminance, carbon dioxide concentration, number of people, and flexible load power.

6. The building flexible intelligent control system for dynamic matching of source and load according to claim 5, characterized in that: The energy storage virtual compartment scheduling coefficient calculation unit is used to calculate the energy storage virtual compartment scheduling coefficient based on the photovoltaic surplus power, peak reduction demand power, comfort guarantee demand power, critical load power and total building power consumption. The second analysis unit is used to compare and analyze the energy storage virtual sub-slot scheduling coefficient with the preset first sub-slot threshold and second sub-slot threshold to obtain a second evaluation result. When the energy storage virtual sub-slot scheduling coefficient reaches the first sub-slot threshold, it is determined that the energy storage system is in a first-level scheduling demand state, and a first sub-slot scheduling strategy is generated. When the energy storage virtual sub-slot scheduling coefficient is between the second sub-slot threshold and the first sub-slot threshold, it is determined that the energy storage system is in a second-level scheduling demand state, and a second sub-slot scheduling strategy is generated. When the energy storage virtual sub-slot scheduling coefficient is lower than the second sub-slot threshold, it is determined that the energy storage system is in a third-level scheduling demand state, and a third sub-slot scheduling strategy is generated.

7. The building flexible intelligent control system for dynamic matching of source and load according to claim 6, characterized in that: The source-load fluctuation suppression assessment module includes a grid connection point power fluctuation calculation unit, a grid connection point voltage offset calculation unit, a photovoltaic output change calculation unit, a building load change calculation unit, a source-load fluctuation suppression coefficient calculation unit, and a third analysis unit; The grid connection point power fluctuation calculation unit is used to calculate the grid connection point power fluctuation coefficient based on the grid connection point active power at the current sampling time, the grid connection point active power at the previous sampling time, and the total building power consumption. The grid connection point voltage offset calculation unit is used to calculate the grid connection point voltage offset coefficient based on the grid connection point voltage and the grid connection point rated voltage. The photovoltaic output change calculation unit is used to calculate the photovoltaic output change coefficient based on the real-time output power of the building photovoltaic system at the current sampling time and the output power of the building photovoltaic system at the previous sampling time. The building load change calculation unit is used to calculate the building load change coefficient based on the total building power consumption at the current sampling time and the total building power consumption at the previous sampling time.

8. A building flexible intelligent control system for dynamic matching of source and load according to claim 7, characterized in that: The source-load fluctuation suppression coefficient calculation unit is used to calculate the source-load fluctuation suppression coefficient based on the grid connection point power fluctuation coefficient, grid connection point voltage deviation coefficient, photovoltaic output variation coefficient, and building load variation coefficient.

9. A building flexible intelligent control system for dynamic matching of source and load according to claim 8, characterized in that: The third analysis unit is used to compare and analyze the source load fluctuation suppression coefficient with the preset first fluctuation threshold and second fluctuation threshold to obtain the third evaluation result; when the source load fluctuation suppression coefficient is lower than the first fluctuation threshold, it is determined that the building source load fluctuation is in an allowable state and a fluctuation allowance strategy is generated. When the source load fluctuation suppression coefficient is between the first fluctuation threshold and the second fluctuation threshold, the building source load fluctuation is determined to be in the first-level suppression state, and a short-term energy storage compensation strategy is generated. When the source load fluctuation suppression coefficient reaches the second fluctuation threshold, the building source load fluctuation is determined to be in the second-level suppression state, and a voltage support and load slow change strategy is generated.

10. A building flexible intelligent control system for dynamic matching of source and load according to claim 9, characterized in that: The flexible control strategy generation module includes a comprehensive strategy determination unit and a final control unit; The comprehensive strategy determination unit is used to generate a final execution strategy based on the first evaluation result, the second evaluation result, and the third evaluation result. When the third evaluation result is a level 2 suppression state, the voltage support and load gradual change strategy is set as the priority execution strategy. When the first evaluation result is a level 2 mismatch state, the forced source-load matching strategy is set as the priority execution strategy. When the second evaluation result is a level 1 scheduling demand state, the first sub-compartment scheduling strategy is set as the priority execution strategy. When the third assessment result is an allowable state, the first assessment result is a matched state, and the second assessment result is a level 3 scheduling demand state, the current operation and maintenance strategy is generated; conflict handling is performed on energy storage charging commands, energy storage discharging commands, energy storage converter voltage support commands, flexible load limiting commands, and flexible load gradual change commands. The final control unit is used to generate final building flexible control instructions based on the conflict handling results, including photovoltaic power supply path control instructions, energy storage compartment capacity call instructions, energy storage charge and discharge control instructions, energy storage converter voltage support instructions, critical load supply guarantee instructions, flexible load limiting instructions, flexible load gradual change instructions, and non-critical load load reduction instructions; and is connected to the building power distribution switchgear, energy storage converter, air conditioning controller, fresh air controller, lighting controller, and sub-item electrical equipment controller respectively to execute the final building flexible control instructions.

Images