A method, system, and equipment for coordinated scheduling of optical and energy storage systems in smart construction sites.

Abstract

This application provides a method, system, and device for coordinated scheduling of photovoltaic and energy storage equipment in smart construction sites, belonging to the field of smart construction site energy management technology. The method includes: acquiring multi-dimensional sensing data of each monitoring sub-region; performing nonlinear correction on photovoltaic output based on the corresponding light intensity, ambient temperature, and construction site shading coefficient of the monitoring sub-region to obtain the corrected photovoltaic output; determining the load demand curve of each monitoring sub-region within a preset time period based on historical load data, real-time operating data, and a load forecasting model; inputting the corrected photovoltaic output, load demand curve, and multi-dimensional sensing data into a dynamic weighted coordination model to determine the output weight triplet; when the real-time fluctuation of the determined output weight triplet exceeds a preset threshold, determining weight adjustment parameters and / or generating a cross-sub-region power supply relay path based on a preset anomaly level matrix, and storing the corresponding scheduling data in a preset database to update the load forecasting model and weight adjustment parameters.

Description

Technical Field

[0001] This application relates to the field of smart construction site energy management technology, and in particular to a method, system and equipment for photovoltaic-storage coordinated scheduling of smart construction site equipment. Background Technology

[0002] Smart construction site equipment (such as tower cranes and concrete mixing plants) generally suffer from problems such as large fluctuations in power load, insufficient stability of traditional power grid supply, and high carbon emissions.

[0003] Photovoltaic energy storage (PV-storage) systems, as a clean energy technology, can supplement construction site electricity through photovoltaic power generation. However, due to factors such as sunlight intensity and fluctuations in equipment energy consumption, low PV-storage synergy efficiency and poor power supply reliability have become prominent pain points. Moreover, existing scheduling methods mostly rely on empirical rules or simple load forecasting, failing to achieve flexible and dynamic precise matching of PV output, energy storage status, and equipment energy consumption.

[0004] Therefore, there is an urgent need for a technical solution that can dynamically and accurately match photovoltaic output, energy storage status and equipment energy consumption. Summary of the Invention

[0005] This application provides a method, system, and device for coordinated scheduling of photovoltaic and energy storage for smart construction site equipment, which solves the technical problem of how to dynamically and accurately match photovoltaic output, energy storage status, and equipment energy consumption to provide stable power supply for smart construction site equipment.

[0006] On one hand, embodiments of this application provide a method for coordinated scheduling of optical and energy storage systems for smart construction site equipment, the method comprising:

[0007] The monitoring sub-areas are divided according to the functional zones and geographical coordinates of the construction site, and multi-dimensional sensing data of each monitoring sub-area is acquired through a pre-deployed sensor network at a corresponding preset monitoring frequency; wherein, the multi-dimensional sensing data includes at least: photovoltaic output, energy storage SOC, equipment energy consumption, and grid stability coefficient.

[0008] Based on the light intensity, ambient temperature and construction site shading coefficient of the monitoring sub-region, the photovoltaic output is nonlinearly corrected to obtain the corrected photovoltaic output.

[0009] Based on the corresponding historical load data, real-time operating condition data, and pre-trained load prediction models, determine the load demand curve of each monitoring sub-area within a preset time period;

[0010] The corrected photovoltaic output, the load demand curve, and the multi-dimensional sensing data are input into a pre-constructed dynamic weighted collaborative model to determine the output weight triplet; the output weight triplet includes photovoltaic output weight, energy storage output weight, and grid output weight.

[0011] When it is determined that the real-time fluctuation of the output weight triplet is greater than a preset threshold, based on the preset anomaly level matrix, the corresponding weight adjustment parameters are determined and / or a cross-sub-region power supply relay path is generated, and the corresponding scheduling data is stored in the preset database to update the load prediction model and the weight adjustment parameters.

[0012] In one implementation of this application, the photovoltaic output is nonlinearly corrected based on the light intensity, ambient temperature, and construction site shading coefficient corresponding to the monitored sub-region to obtain the corrected photovoltaic output, specifically including:

[0013] Calculate the light intensity correction parameters based on the light intensity and the light nonlinearity correction coefficient;

[0014] Calculate the ambient temperature correction parameters based on the ambient temperature and the preset temperature correction coefficient;

[0015] The corrected photovoltaic output is calculated based on the photovoltaic output, the light intensity correction parameter, the ambient temperature correction parameter, and the construction site shading coefficient.

[0016] In one implementation of this application, before calculating the corrected photovoltaic output based on the photovoltaic output, the light intensity correction parameter, the ambient temperature correction parameter, and the construction site shading coefficient, the method further includes:

[0017] Obtain the real-time location coordinates of the tower cranes and the photovoltaic arrays at the smart construction site, and unify them into the global coordinate system of the construction site;

[0018] Obtain the real-time solar altitude angle and solar azimuth angle from the user terminal, and calculate the unit vector of the solar direction;

[0019] Based on the real-time location coordinates, the solar direction unit vector, and the preset shadow projection model, the tower shadow area and the crane boom shadow area corresponding to the smart construction site tower crane are determined.

[0020] Based on the coordinates of the photovoltaic array, the shadow area of ​​the tower body, and the shadow area of ​​the crane arm, the shaded area corresponding to the photovoltaic array is determined, and the site shading coefficient corresponding to the photovoltaic array is determined based on the shaded area and the total area of ​​the photovoltaic array.

[0021] In one implementation of this application, before determining the load demand curve of each monitoring sub-region within a preset time period based on corresponding historical load data, real-time operating condition data, and a pre-trained load prediction model, the method further includes:

[0022] Determine the regional type of the monitored sub-region corresponding to the load demand curve to be predicted; the regional type at least includes a hoisting area, a mixing area, and an auxiliary area; the monitored sub-region corresponding to the hoisting area at least includes the following intelligent construction site equipment: tower cranes, construction elevators; the monitored sub-region corresponding to the auxiliary area at least includes the following intelligent construction site equipment: lighting equipment, monitoring equipment;

[0023] According to the regional type, determine the corresponding model channels to call the first prediction sub-model in the load prediction model to perform load prediction on the monitored sub-region of the hoisting area, and call the second prediction sub-model in the load prediction model to perform load prediction on the monitored sub-region of the auxiliary area; the required computing power of the first prediction sub-model is at least greater than that of the second prediction sub-model.

[0024] In an implementation manner of the present application, input the corrected photovoltaic output power, the load demand curve, and the energy storage SOC into a pre-constructed dynamic weight collaboration model to determine the output power weight triple, specifically including:

[0025] Taking meeting the load demand curve as the goal, through the dynamic weight collaboration model, according to the corrected photovoltaic output power, the maximum photovoltaic output power, the grid stability coefficient, the loads of each device in the corresponding monitored sub-region, and the device priority coefficients, determine the corresponding photovoltaic output power weight; wherein, the device priority coefficient is positively correlated with the device energy consumption;

[0026] Through the dynamic weight collaboration model, according to the energy storage SOC, the loads of each device, the device priority coefficients, the corrected photovoltaic output power, and the maximum energy storage output power, determine the corresponding energy storage output power weight;

[0027] Through the dynamic weight collaboration model, according to the photovoltaic output power weight and the energy storage output power weight, determine the grid output power weight, and add the photovoltaic output power weight, the energy storage output power weight, and the grid output power weight to the output power weight triple. [[ID=​​​​​​Determine the real-time fluctuation amplitude and corresponding fluctuation duration of the output weight triplet;

[0031] The real-time fluctuation amplitude is compared with each preset fluctuation amplitude in the preset anomaly level matrix, and the fluctuation duration is compared with each preset fluctuation duration in the preset anomaly level matrix, so as to determine the anomaly level based on the comparison results.

[0032] When the anomaly level is Level 1, the corresponding weight adjustment parameters are determined according to the preset temporary suppression correction reference table to correct each output weight in the output weight triplet; the correction includes positive correction and negative correction.

[0033] When the anomaly level is level two, the corresponding weight adjustment parameters are determined according to the preset temporary suppression correction reference table, the local output weight in the output weight triplet is corrected, and the cross-sub-region power supply relay path is generated so that other monitoring sub-regions can supply power to the corresponding monitoring sub-region.

[0034] When the anomaly level is level three, the corresponding weight adjustment parameters are determined according to the preset temporary suppression correction reference table, the global output weight in the output weight triplet is corrected, and the cross-sub-region power supply relay path is generated so that other monitoring sub-regions can supply power to the corresponding monitoring sub-region.

[0035] In one implementation of this application, generating a cross-sub-region power relay path specifically includes:

[0036] Construct a smart construction site topology map, wherein the nodes of the construction site topology map are each of the monitoring sub-regions, and the edges are power transmission lines;

[0037] Based on the energy storage SOC and the maximum preset energy storage output of each node, the energy redundancy corresponding to each node is determined.

[0038] The transmission loss corresponding to each side is determined based on the resistivity of the transmission line, the distance between each side, and the preset cross-sectional area of ​​the conductor.

[0039] The abnormal monitoring sub-regions where the real-time fluctuation of the output weight triplet is greater than a preset threshold are taken as target support regions. Based on the load gap value corresponding to the target support region and the energy redundancy corresponding to each monitoring sub-region, one or more candidate support regions are screened in each monitoring sub-region. When there is only one candidate support region, it is determined as the support region.

[0040] When there are multiple candidate support regions, each path weight is determined based on the energy redundancy and corresponding transmission loss of each candidate support region, so as to select a support region according to the path weight; wherein the sum of the energy redundancy corresponding to multiple support regions is at least greater than the load gap value of the target support region;

[0041] Based on the edge between the support area and the target support area in the construction site topology map, the cross-sub-region power supply relay path is established.

[0042] In one implementation of this application, the supported regions are selected based on the path weights, specifically including:

[0043] If multiple target support regions are determined to exist, the product of the device priority coefficient corresponding to the target support region and the corresponding path weight is calculated, and the product value is used as the updated path weight to filter out the support regions according to the updated path weight.

[0044] Secondly, embodiments of this application also provide a photoelectric storage collaborative scheduling system for smart construction site equipment, the system comprising:

[0045] The acquisition module is used to divide the monitoring sub-areas according to the functional zoning of the construction site and the geographical coordinates, and to acquire multi-dimensional sensing data of each monitoring sub-area through a pre-deployed sensor network at a corresponding preset monitoring frequency; wherein, the multi-dimensional sensing data includes at least: photovoltaic output, energy storage SOC, equipment energy consumption, and grid stability coefficient.

[0046] The correction module is used to perform nonlinear correction on the photovoltaic output based on the light intensity, ambient temperature and construction site shading coefficient corresponding to the monitoring sub-region, so as to obtain the corrected photovoltaic output.

[0047] The first determining module is used to determine the load demand curve of each monitoring sub-region within a preset time period based on the corresponding historical load data, real-time operating condition data and pre-trained load prediction model.

[0048] The input determination module is used to input the corrected photovoltaic output, the load demand curve, and the multi-dimensional sensing data into a pre-constructed dynamic weighted coordination model to determine the output weight triplet; the output weight triplet includes photovoltaic output weight, energy storage output weight, and grid output weight.

[0049] The second determining module is used to determine the corresponding weight adjustment parameters and / or generate a cross-sub-region power supply relay path based on a preset anomaly level matrix when the real-time fluctuation of the output weight triplet is determined to be greater than a preset threshold, and to store the corresponding scheduling data in a preset database to update the load prediction model and the weight adjustment parameters.

[0050] Thirdly, this application also provides a photoelectric storage collaborative scheduling device for smart construction site equipment, the device comprising:

[0051] At least one processor; and a memory communicatively connected to the at least one processor; wherein the memory stores instructions executable by the at least one processor, the instructions being executed by the at least one processor to enable the at least one processor to perform a photoelectric storage collaborative scheduling method for smart construction site equipment as described above.

[0052] Compared with the prior art, the significant advantages of this application are as follows:

[0053] The above technical solution enables data sensing and monitoring of photovoltaic (PV) and energy storage in monitoring sub-regions with different functions and locations. It allows for adaptive correction of PV output and accurate prediction of load demand. Subsequently, it determines the dynamic weights of PV, energy storage, and the power grid, achieving precise and coordinated allocation of these dynamic weights for more flexible PV-energy storage coordinated scheduling. Furthermore, this application establishes an anomaly response mechanism, triggering weight adjustment parameters or cross-sub-region power supply relay paths based on anomaly level matrices, avoiding the problems of one-size-fits-all or delayed anomaly responses in traditional methods. This achieves dynamic and precise matching of PV output, energy storage status, and equipment energy consumption, enabling intelligent PV-energy storage coordinated scheduling and providing stable power supply for smart construction site equipment. Attached Figure Description

[0054] The accompanying drawings, which are included to provide a further understanding of this application and form part of this application, illustrate exemplary embodiments and are used to explain this application, but do not constitute an undue limitation of this application. In the drawings:

[0055] Figure 1 This is a flowchart illustrating a method for coordinated scheduling of optical and energy storage systems for smart construction site equipment, as described in an embodiment of this application.

[0056] Figure 2 This is a schematic diagram of the structure of a photoelectric storage collaborative scheduling system for smart construction site equipment in an embodiment of this application;

[0057] Figure 3 This is a schematic diagram of the structure of a light-storage collaborative scheduling device for smart construction site equipment in an embodiment of this application. Detailed Implementation

[0058] To make the objectives, technical solutions, and advantages of this application clearer, the technical solutions of this application will be clearly and completely described below in conjunction with specific embodiments and corresponding drawings. Obviously, the described embodiments are only a part of the embodiments of this application, and not all of them. Based on the embodiments in this application, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of this application.

[0059] Smart construction site equipment (such as tower cranes and concrete mixing plants) generally suffers from problems such as large fluctuations in electricity load, insufficient stability of traditional power grid supply, and high carbon emissions. Photovoltaic energy storage (PV-storage) systems, as a clean energy technology, can supplement construction site electricity through photovoltaic power generation. However, due to factors such as sunlight intensity and fluctuations in equipment energy consumption, low PV-storage synergy efficiency and poor power supply reliability have become prominent pain points. Furthermore, existing scheduling methods mostly rely on empirical rules or simple load forecasting, failing to achieve flexible and dynamic precise matching between photovoltaic output, energy storage status, and equipment energy consumption.

[0060] Based on this, embodiments of this application provide a method, system, and device for coordinated scheduling of photovoltaic and energy storage for smart construction site equipment, which solves the technical problem of how to dynamically and accurately match photovoltaic output, energy storage status, and equipment energy consumption to provide stable power supply for smart construction site equipment.

[0061] The various embodiments of this application are described in detail below with reference to the accompanying drawings.

[0062] This application provides a method for coordinated scheduling of optical and energy storage systems for smart construction site equipment, such as... Figure 1 As shown, the method may include steps S101-S105:

[0063] S101, the server divides the monitoring sub-areas according to the functional zones of the construction site and the geographical coordinates, and obtains multi-dimensional sensing data of each monitoring sub-area through a pre-deployed sensor network at a corresponding preset monitoring frequency.

[0064] The multi-dimensional sensing data includes at least: photovoltaic output, remaining energy storage capacity (State of Charge, SOC), equipment energy consumption, and grid stability coefficient.

[0065] The construction site can be functionally divided according to area type, such as hoisting area, mixing area, auxiliary area, photovoltaic array area, living area, etc. The specific division can be set by the user based on the actual smart construction site, or by expert experience; no specific limitations are imposed here. To avoid overlapping or omissions in monitoring areas, the spatial boundaries of each sub-area will be clearly defined using geographic coordinates, thus forming an electronic map of the monitoring sub-areas distribution within the smart construction site. Subsequently, a sensor network pre-deployed in each monitoring sub-area will be used to collect data. This sensor network may include sensors such as GPS positioning sensors, tilt sensors, temperature and humidity sensors, current sensors, irradiance sensors, power sensors, voltage sensors, and harmonic analysis sensors; the specific configuration is determined by the user based on the specific needs of the smart construction site, and no specific limitations are imposed here.

[0066] The photovoltaic output can be collected using power sensors; the energy storage SOC can be calculated using voltage and current sensors, and then a preset SOC calculation tool can be used, with the specific settings determined by the user based on the actual usage scenario; equipment energy consumption can be obtained using current and voltage sensors, and then an energy consumption calculation tool can be used; the grid stability coefficient reflects the stability of grid voltage, frequency, harmonics, and other indicators, and can be calculated using voltage, frequency, and harmonic analysis sensors, and then an expert system can be used to collect data. The photovoltaic output, energy storage SOC, equipment energy consumption, and grid stability coefficient can all be obtained by inputting the collected data into the expert system, and this application does not impose specific limitations on these parameters.

[0067] Furthermore, this application will pre-set preset monitoring frequencies for different monitoring sub-regions to flexibly collect multi-dimensional sensing data for each monitoring sub-region. By acquiring data at differentiated collection time intervals using preset monitoring frequencies, data redundancy or information lag can be avoided, effectively ensuring data timeliness.

[0068] It should be noted that the server, as the executing entity of the optical-storage collaborative scheduling method for smart construction site equipment, is only an example. The executing entity is not limited to the server, but can also be an edge computing terminal set up at the smart construction site. This application does not make any specific limitations in this regard.

[0069] S102, the server performs nonlinear correction on the photovoltaic output based on the light intensity, ambient temperature and construction site shading coefficient of the monitored sub-region, and obtains the corrected photovoltaic output.

[0070] In this embodiment of the application, the photovoltaic output is nonlinearly corrected based on the light intensity, ambient temperature, and construction site shading coefficient corresponding to the monitored sub-region, to obtain the corrected photovoltaic output, specifically including:

[0071] Calculate the illuminance correction parameters based on the illuminance intensity and the illuminance nonlinearity correction coefficient. Calculate the ambient temperature correction parameters based on the ambient temperature and the preset temperature correction coefficient. Calculate the corrected photovoltaic output based on the photovoltaic output, illuminance correction parameters, ambient temperature correction parameters, and site shading coefficient.

[0072] The specific formula for nonlinear correction of photovoltaic power output in this application is as follows:

[0073] .

[0074] in, Indicates the first Corrected photovoltaic output for each monitoring sub-region Indicates the first The original photovoltaic output of each monitored sub-region This indicates the preset temperature correction factor. Indicates the first The ambient temperature of each monitoring sub-region This represents the preset illumination nonlinearity correction coefficient. Indicates light intensity. Indicates the first The construction site occlusion coefficient at any given time. This represents the light intensity correction parameter; This indicates the ambient temperature correction parameter.

[0075] The above formula can be used to adjust the photovoltaic output by taking into account factors such as light intensity, temperature, and site-specific shading, thus obtaining an accurate photovoltaic output.

[0076] Since the location of the tower crane and the angle of the sun are constantly changing, the construction site shading coefficient is not an absolute value. Therefore, before calculating the corrected photovoltaic output based on the photovoltaic output, light intensity correction parameters, ambient temperature correction parameters, and the construction site shading coefficient, this application can calculate the construction site shading coefficient, including:

[0077] The system acquires the real-time location coordinates of the tower crane and the photovoltaic array at the smart construction site, and unifies them into the global coordinate system of the construction site. It also acquires the real-time solar altitude angle and solar azimuth angle from the user terminal and calculates the solar direction unit vector. Based on the real-time location coordinates, the solar direction unit vector, and a preset shadow projection model, it determines the shadow areas of the tower crane and the boom. According to the photovoltaic array coordinates, the tower shadow area, and the boom shadow area, it determines the shaded area of ​​the photovoltaic array. Based on the shaded area and the total area of ​​the photovoltaic array, it determines the site shading coefficient for the corresponding photovoltaic array.

[0078] In other words, the server can obtain the real-time position coordinates of the tower crane and the photovoltaic array coordinates at the smart construction site through sensors installed on the tower crane or information from user terminals. The real-time position coordinates include at least the tower crane coordinates and the coordinates of the boom end point. :

[0079] .

[0080] in, This indicates the coordinates of the tower crane corresponding to the boom. The length of the crane boom. The azimuth angle of the crane boom can be a rotation angle relative to a predetermined direction, such as due north. This is the elevation angle of the crane boom, i.e., the angle with the horizontal plane. The coordinates of the photovoltaic array are... , The photovoltaic array represents the first When the photovoltaic array is a rectangular array, the photovoltaic array coordinates must include the coordinates of all four vertices. This application unifies the above coordinates into the global coordinate system of the construction site to facilitate the calculation of shadow occlusion.

[0081] The solar altitude angle and solar azimuth angle can be obtained from a user terminal, which can be a user's mobile phone, computer, or other device. The user terminal can connect to other sensors or the Internet to obtain the solar altitude angle at the current time and latitude of the smart construction site. Azimuth Then calculate the unit vector in the direction of the sun. The solar altitude angle at the horizon is 0 degrees, with a range of 0 to 90 degrees. The solar azimuth angle in the due south direction is 0 degrees, while the solar azimuth angle in the west direction is positive, with a range of -180° to 180°.

[0082] Subsequently, the server can call a preset shadow projection model to calculate the shadow area. Specifically, the preset shadow projection model treats the tower crane body as a cylinder with a radius of... The height is Its projection on the ground is abstracted as an ellipse, and the coordinates of its shadow center are... : , Long axis short axis This yields the shadow area of ​​the tower. For the crane boom, its shadow is a band-shaped area formed on the ground along a shadow line segment, which is defined by... Starting from the shaded coordinates As the endpoint, The calculation formula is as follows:

[0083] .

[0084] The server is based on the preset length and width. To obtain the width of the shadow on the ground. The shaded area of ​​the crane boom is along both sides of the shaded line segment. A strip-shaped area.

[0085] Subsequently, the server will calculate whether the photovoltaic array coordinates are within the shadow areas of the tower and the crane arm. This can be determined by calculating the distance from the apex of the photovoltaic array to the center point of the shadow area, or by using a pre-trained neural network model to determine whether the photovoltaic array area overlaps with the shadow area. This application does not specifically limit this method. When the photovoltaic array area overlaps with the shadow area, the area of ​​the overlapping area will be calculated. This area is the shaded area. This application can output the shaded area using the aforementioned neural network model, or it can use image acquisition and other methods to identify the shadow area and calculate the shaded area using computer vision algorithms. This algorithm can be deployed in image acquisition equipment, and this application does not specifically limit this method.

[0086] Subsequently, the server followed the formula The construction site shading coefficient was calculated, where, This indicates the preset attenuation coefficient, which is set by the user based on expert experience. This represents the sum of the overlapping areas of the photovoltaic array area and the shadow areas of the tower and crane arm. This represents the total area of ​​the photovoltaic array.

[0087] S103, the server determines the load demand curve of each monitoring sub-area within a preset time period based on the corresponding historical load data, real-time operating condition data and pre-trained load prediction model.

[0088] In this embodiment of the application, before determining the load demand curve of each monitoring sub-region within a preset time period based on the corresponding historical load data, real-time operating condition data, and pre-trained load prediction model, the method further includes:

[0089] Determine the region type of the monitoring sub-region corresponding to the load demand curve to be predicted. The region type must include at least three categories: hoisting area, mixing area, and auxiliary area. The monitoring sub-region corresponding to the hoisting area must include at least the following smart construction site equipment: tower cranes and construction elevators. The monitoring sub-region corresponding to the auxiliary area must include at least the following smart construction site equipment: lighting equipment and monitoring equipment. Based on the region type, determine the corresponding model channel to call the first prediction sub-model in the load prediction model to perform load prediction for the hoisting area monitoring sub-region, and to call the second prediction sub-model in the load prediction model to perform load prediction for the auxiliary area monitoring sub-region. The required computing power of the first prediction sub-model must be at least greater than that of the second prediction sub-model.

[0090] In other words, different sub-models from different load forecasting models can be used for forecasting different region types. The first forecasting sub-model can be a pre-trained Long Short-Term Memory (LSTM) network model, while the second forecasting sub-model can be a model with lower computational requirements, such as the ARIMA model or exponential smoothing. The specific model algorithm used can also be adjusted according to the actual application scenario, and this application does not impose specific limitations on it. By selecting different models to process data from different region types and perform load forecasting, accurate forecasting is achieved for important regions, while load forecasting for ordinary regions maintains computational efficiency, avoiding wasted computing power and improving the data processing efficiency of optical-storage collaborative scheduling.

[0091] The aforementioned historical load data can be pre-stored in a database connected to the server. When executing the load demand curve, the server can retrieve the historical load data from this database. Real-time operating condition data, such as temperature, humidity, light intensity, number of operating devices, and device power ratings for each monitoring sub-area, are specifically set by the user according to the scenario and are not specifically limited here. By inputting historical load data and real-time operating condition data over a period of time into the load forecasting model, the load forecasting model can output a load demand curve for a preset time period.

[0092] S104, the server inputs the corrected photovoltaic output, load demand curves and multi-dimensional sensing data into the pre-built dynamic weighted collaborative model to determine the output weight triplet.

[0093] The output weighting ternary group includes photovoltaic output weighting, energy storage output weighting, and grid output weighting.

[0094] In this embodiment, the modified photovoltaic output, load demand curves, and energy storage SOC are input into a pre-built dynamic weighted collaborative model to determine the output weight triplet, specifically including:

[0095] To meet the load demand curve, a dynamic weighted coordination model is used to determine the corresponding photovoltaic (PV) output weights based on the corrected PV output, maximum PV output, grid stability coefficient, load of each device in the corresponding monitoring sub-region, and priority coefficients of each device. The device priority coefficient is positively correlated with device energy consumption. The same dynamic weighted coordination model is used to determine the corresponding energy storage output weights based on the energy storage SOC, load of each device, priority coefficients of each device, corrected PV output, and maximum energy storage output. Finally, the dynamic weighted coordination model is used to determine the grid output weights based on the PV output weights and energy storage output weights. These PV output weights, energy storage output weights, and grid output weights are then added to the output weight triplet.

[0096] Specifically, the dynamic weighted coordination model uses the load demand value at each moment in the aforementioned load demand curve as the target, and calculates the output weights for the three dimensions of photovoltaics, energy storage, and the power grid. This involves multiplying the adjusted output weights by the processing power of each of the three dimensions, and comparing the sum of these products with the corresponding load demand value in the load demand curve. "Corresponding" refers to the moment when the two comparison values ​​are at the same point in time. If at least the sum of the products of each dimension is greater than the load demand value in the load demand curve, then the output weights are considered to meet the requirements. Otherwise, the calculation and updating of the output weights for the three dimensions continues until the sum of the products of each dimension is greater than the load demand value in the load demand curve.

[0097] The formula for calculating the photovoltaic output weight is as follows:

[0098] .

[0099] in, Indicates the first The photovoltaic output weight of each monitoring sub-region For the first The maximum photovoltaic output of each monitored sub-region Represents the power grid stability coefficient. Indicates the first Equipment priority coefficient for class of equipment Indicates the first Equipment load of this type of equipment This indicates the total number of devices in different categories, which can be classified according to their energy consumption.

[0100] The formula for calculating the energy storage output weight is as follows:

[0101] .

[0102] in, Indicates the first Energy storage output weight of each monitoring sub-region Indicates the State of Charge (SOC) for energy storage. Indicates the first The maximum energy storage output preset for each monitoring sub-area.

[0103] The power grid output weight is: .

[0104] The aforementioned photovoltaic (PV) output weight, energy storage output weight, and grid output weight are used to adjust the PV output, energy storage output, and grid output, respectively. PV output, energy storage output, and grid output can be understood as power, with units of kilowatts (kW). After obtaining the output weight tripartite, which includes the PV output weight, energy storage output weight, and grid output weight, the maximum output power of the PV system can be corrected, the discharge threshold and charging / discharging current of the energy storage system can be adjusted, and the maximum power of the grid can be adjusted. That is, this application can specifically adjust the scheduling parameters in the current PV-energy storage coordinated scheduling strategy based on the output weight tripartite. The specific adjustment method can be obtained from expert experience, and this application does not impose specific limitations on it.

[0105] S105 When the server determines that the real-time fluctuation of the output weight triplet is greater than the preset threshold, it determines the corresponding weight adjustment parameters and / or generates a cross-sub-region power supply relay path based on the preset anomaly level matrix, and stores the corresponding scheduling data in the preset database to update the load prediction model and weight adjustment parameters.

[0106] The server can statistically analyze the real-time fluctuations of the output weight triplet within a preset fluctuation detection time window, and calculate the continuous change in output weight within that window. Specifically, this continuous change in output weight can be obtained by differentiating the weight change curves of each output weight within the fluctuation detection time window and summing the absolute values ​​of the three derivatives. This continuous change in output weight is then used as the fluctuation value corresponding to the real-time fluctuation. Subsequently, the fluctuation value of this real-time fluctuation is compared with a preset threshold. This preset threshold can be set by the user or obtained based on expert experience, and it can be updated periodically based on expert experience; this application does not specifically limit its application in this regard.

[0107] In this embodiment of the application, based on a preset anomaly level matrix, the corresponding weight adjustment parameters are determined and / or a cross-sub-region power supply relay path is generated, specifically including:

[0108] Preset a first preset fluctuation amplitude F1, a second preset fluctuation amplitude F2, and a third preset fluctuation amplitude F3, and F1 < F2 < F3. Preset a first preset fluctuation duration t1 and a second preset fluctuation duration t2, where t1 < t2. Then, construct a preset abnormal level matrix according to the abnormal level preset by the user and the combinations of each preset fluctuation amplitude and each preset fluctuation duration. Determine the real-time fluctuation amplitude corresponding to the output power weight triple and the corresponding fluctuation duration. Compare the real-time fluctuation amplitude with each preset fluctuation amplitude in the preset abnormal level matrix, and compare the fluctuation duration with each preset fluctuation duration in each preset abnormal level matrix to determine the abnormal level according to the comparison results. When the abnormal level is level one, determine the corresponding weight adjustment parameter according to the preset temporary wave suppression correction comparison table to correct each output power weight in the output power weight triple. The correction includes positive correction and negative correction. When the abnormal level is level two, determine the corresponding weight adjustment parameter according to the preset temporary wave suppression correction comparison table, correct the local output power weight in the output power weight triple, and generate a cross-sub-region power supply relay path to enable other monitoring sub-regions to supply power to the corresponding monitoring sub-region. When the abnormal level is level three, determine the corresponding weight adjustment parameter according to the preset temporary wave suppression correction comparison table, correct the global output power weight in the output power weight triple, and generate a cross-sub-region power supply relay path to enable other monitoring sub-regions to supply power to the corresponding monitoring sub-region. The preset temporary wave suppression correction comparison table can be set by the user according to the actual usage scenario, and includes the corresponding relationship between different abnormal levels and the correction step size of the weight adjustment parameter. The corresponding correction step size can be obtained through the preset temporary wave suppression correction comparison table to perform correction processing on the weight adjustment parameter.

[0109] In other words, this application presets a preset abnormal level matrix including the corresponding relationship between the preset fluctuation amplitude, the preset fluctuation duration, and the abnormal level. The preset abnormal level matrix represents the first preset fluctuation amplitude F1, the second preset fluctuation amplitude F2, and the third preset fluctuation amplitude F3, as well as the first preset fluctuation duration t1 and the second preset fluctuation duration t2 through different coding values. The server will further determine the real-time fluctuation amplitude of the output power weight triple and the fluctuation duration. Subsequently, the abnormal level is determined by comparing the real-time fluctuation amplitude and the fluctuation duration with the preset abnormal level matrix. The comparison results are shown in Table 1 below.

[0110] Table 1 Comparison Result Table

[0111]

[0112] For different abnormal levels, this application will adopt different abnormal handling methods to flexibly handle different abnormal fluctuations, so as to make the photovoltaic-storage collaborative scheduling more flexible.

[0113] It should be noted that the aforementioned local output weight can be understood as a portion of the output weight in the output weight triplet, while the global output weight represents all the output weights in the output weight triplet.

[0114] Furthermore, there are differences in the generation of cross-sub-region power supply relay paths for anomaly levels two and three. For example, in the case of a level two anomaly, the path is preferentially selected from other monitored sub-regions with high energy redundancy and short distance. In the case of a level three anomaly, the path is preferentially selected from other monitored sub-regions with large transmission capacity (e.g., conductor cross-sectional area greater than 50 square millimeters) and high reliability (e.g., historical transmission failure rate less than 1%). Specific settings can be configured by the user according to the actual usage scenario, and are not specifically limited here.

[0115] In one embodiment of this application, the above-mentioned generation of a cross-sub-region power supply relay path specifically includes:

[0116] A smart construction site topology map is constructed, where nodes represent monitoring sub-regions and edges represent transmission lines. The energy redundancy of each node is determined based on its State of Charge (SOC) and the maximum preset energy storage output. Transmission losses for each edge are determined based on the resistivity of the transmission lines, the distance between edges, and the preset cross-sectional area of ​​the conductors. Abnormal monitoring sub-regions with real-time fluctuations exceeding a preset threshold in the output weight triplet are designated as target support regions. Based on the load gap value corresponding to the target support region and the energy redundancy of each monitoring sub-region, one or more candidate support regions are selected from each monitoring sub-region. If there is only one candidate support region, it is designated as the support region. If multiple candidate support regions exist, path weights are determined based on the energy redundancy and corresponding transmission losses of each candidate support region, and support regions are selected based on these path weights. The sum of the energy redundancies of multiple support regions must be at least greater than the load gap value of the target support region. A cross-sub-region power supply relay path is established based on the edges between the support regions and the target support region in the construction site topology map.

[0117] In other words, this application first constructs a site topology map, then calculates energy redundancy and transmission loss, and finally calculates path weights to generate the shortest power supply path from the support area to the target support area that needs support. Energy redundancy is determined through... Calculations show that Indicates the first Energy storage SOC of each monitoring sub-region Indicates the first The energy redundancy of each monitoring sub-region. Transmission loss at the edges is... Calculations show that Indicates the first The monitoring sub-region and the first Transmission loss of edges between monitoring sub-regions, Represents resistivity. Indicates the first The monitoring sub-region and the first The distance between the edges of the monitored sub-regions Indicates the first The monitoring sub-region and the first Preset cross-sectional area of ​​transmission line conductors between monitoring sub-areas.

[0118] Subsequently, the server will further calculate the actual load demand at this time through the load forecasting model, and calculate the actual output power (output) of photovoltaic, energy storage and grid respectively. The difference between the actual load demand and the actual output power will be calculated to obtain the load gap value.

[0119] Before screening one or more candidate support areas in each monitoring sub-region based on the load gap value corresponding to the target support area and the energy redundancy corresponding to each monitoring sub-region, if the load gap value is negative, it indicates an energy supply surplus. In this case, the label of the target support area is changed to a candidate support area for screening. If the load gap value is positive, it indicates an energy supply shortage. In this case, it is used as the target support area.

[0120] The server then compares the load gap value with the energy redundancy of other monitored sub-regions, filtering out one or more candidate support regions. The sum of the energy redundancy of the candidate support regions must be greater than the load gap value. Additionally, if there is only one candidate support region, then the energy redundancy of that candidate support region itself must be greater than the load gap value.

[0121] If there are multiple candidate support regions, then it will be passed through The path weights are calculated. Indicates the first The monitoring sub-region and the first The path weights of each monitoring sub-region are determined by ranking them from largest to smallest, and support regions are selected while ensuring that the sum of the energy redundancy of the support regions is at least greater than the load gap value. Then, based on the selected support regions and the site topology map, a cross-sub-region power supply relay path is established, generating a path map that includes power supply from other selected monitoring sub-regions to the target support region.

[0122] Furthermore, since smart construction sites may involve simultaneous construction at multiple locations, there may be multiple target support areas. Therefore, this application selects support areas based on path weights, specifically including:

[0123] If multiple target support areas are identified, the product of the device priority coefficient corresponding to the target support area and the corresponding path weight is calculated, and the product value is used as the updated path weight to filter out the support areas based on the updated path weight.

[0124] In other words, if there are multiple target support areas in this application, a preemption conflict may occur when selecting support areas. In this case, this application adds a device priority coefficient, calculates the product of the path weight and the device priority coefficient, and further adds priority factors related to the target support area to the support area, so as to prioritize the support and replenishment of power to the target support area with a larger updated path weight, thereby meeting the load gap of the target support area.

[0125] In addition, this application will store the scheduling data in a preset database electrically connected to the server, so as to serve as historical data for retraining the load forecasting model and for experts or users to update the weight adjustment parameters.

[0126] This application, through the aforementioned technical solution, enables data sensing and monitoring of photovoltaic (PV) and energy storage (ESD) sub-regions with different functions and locations. It allows for adaptive correction of PV output and accurate prediction of load demand. Subsequently, it determines the dynamic weights of PV, ESD, and the grid, achieving precise and coordinated allocation of these dynamic weights for more flexible PV-ESD coordinated scheduling. Furthermore, this application establishes an anomaly response mechanism, triggering weight adjustment parameters or cross-sub-region power supply relay paths based on anomaly level matrices, avoiding the problems of one-size-fits-all or delayed anomaly responses found in traditional methods. This results in dynamic and precise matching of PV output, ESD status, and equipment energy consumption, enabling intelligent PV-ESD coordinated scheduling and providing stable power supply for smart construction site equipment.

[0127] This application, through a full-process design of "precise perception - intelligent correction - dynamic allocation - anomaly coordination - closed-loop optimization", significantly improves the accuracy, stability and economy of smart construction site photovoltaic-storage collaborative scheduling, and provides reliable technical support for the efficient use of energy in complex construction site scenarios.

[0128] Figure 2 A schematic diagram of a light-storage collaborative scheduling system for smart construction site equipment provided in this application embodiment is shown below. Figure 2 As shown, the optical-storage collaborative scheduling system 200 for smart construction site equipment includes:

[0129] The acquisition module 201 is used to divide the monitoring sub-areas according to the functional zoning and geographical coordinates of the construction site, and acquire multi-dimensional sensing data of each monitoring sub-area through a pre-deployed sensor network at a corresponding preset monitoring frequency. The multi-dimensional sensing data includes at least: photovoltaic output, energy storage SOC, equipment energy consumption, and grid stability coefficient. The correction module 202 is used to perform nonlinear correction on the photovoltaic output based on the corresponding light intensity, ambient temperature, and construction site shading coefficient of the monitoring sub-area, obtaining the corrected photovoltaic output. The first determination module 203 is used to determine the load demand curve of each monitoring sub-area within a preset time period based on the corresponding historical load data, real-time operating condition data, and a pre-trained load prediction model. The input determination module 204 is used to input the corrected photovoltaic output, load demand curve, and multi-dimensional sensing data into a pre-constructed dynamic weighted collaborative model to determine the output weight triplet. The output weight triplet includes photovoltaic output weight, energy storage output weight, and grid output weight. The second determining module 205 is used to determine the corresponding weight adjustment parameters and / or generate a cross-sub-region power supply relay path based on a preset anomaly level matrix when the real-time fluctuation of the determined output weight triplet is greater than a preset threshold, and to store the corresponding scheduling data in a preset database to update the load forecasting model and weight adjustment parameters.

[0130] Figure 3 A schematic diagram of a light-storage collaborative scheduling device for smart construction site equipment provided in this application embodiment is shown below. Figure 3 As shown, the device includes:

[0131] At least one processor; and a memory communicatively connected to the at least one processor. The memory stores instructions executable by the at least one processor, which, when executed by the at least one processor, enable the at least one processor to:

[0132] The monitoring sub-regions are divided according to the functional zoning and geographical coordinates of the construction site. Multidimensional sensing data for each sub-region is acquired through a pre-deployed sensor network at a predetermined monitoring frequency. This multidimensional sensing data includes at least: photovoltaic (PV) output, energy storage SOC, equipment energy consumption, and grid stability coefficient. Based on the corresponding irradiance, ambient temperature, and site shading coefficient of the monitoring sub-region, the PV output is nonlinearly corrected to obtain the corrected PV output. Based on historical load data, real-time operating data, and a pre-trained load forecasting model, the load demand curve for each monitoring sub-region within a predetermined time period is determined. The corrected PV output, load demand curve, and multidimensional sensing data are input into a pre-constructed dynamic weighted collaborative model to determine the output weight triplet. The output weight triplet includes the PV output weight, energy storage output weight, and grid output weight. When the real-time fluctuation of the output weight triplet exceeds a predetermined threshold, based on a predetermined anomaly level matrix, corresponding weight adjustment parameters are determined and / or a cross-sub-region power supply relay path is generated. The corresponding scheduling data is then stored in a predetermined database to update the load forecasting model and weight adjustment parameters.

[0133] The various embodiments in this application are described in a progressive manner. Similar or identical parts between embodiments can be referred to mutually. Each embodiment focuses on describing the differences from other embodiments. In particular, the system and device embodiments are basically similar to the method embodiments, so the description is relatively simple; relevant parts can be referred to the description of the method embodiments.

[0134] The systems, devices, and methods provided in this application are one-to-one correspondences. Therefore, the systems and devices also have similar beneficial technical effects as their corresponding methods. Since the beneficial technical effects of the methods have been described in detail above, the beneficial technical effects of the systems and devices will not be repeated here.

[0135] It should also be noted that the terms "comprising," "including," or any other variations thereof are intended to cover non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements includes not only those elements but also other elements not expressly listed, or elements inherent to such a process, method, article, or apparatus. Without further limitation, an element defined by the phrase "comprising one..." does not exclude the presence of other identical elements in the process, method, article, or apparatus that includes said element.

[0136] The above description is merely an embodiment of this application and is not intended to limit the scope of this application. Various modifications and variations can be made to this application by those skilled in the art. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of this application should be included within the scope of the claims of this application.

Claims

1. A method for coordinated scheduling of optical and energy storage systems for smart construction site equipment, characterized in that, The method includes: Dividing the monitoring sub - regions according to the construction site functional areas and geographical location coordinates, and obtaining the multi - dimensional perception data of each monitoring sub - region through a pre - deployed sensor network at a corresponding preset monitoring frequency; wherein, the multi - dimensional perception data at least includes: photovoltaic power output, energy storage SOC, equipment energy consumption, grid stability coefficient; Based on the light intensity, ambient temperature and construction site occlusion coefficient corresponding to the monitoring sub - region, non - linearly correcting the photovoltaic power output to obtain the corrected photovoltaic power output; According to the corresponding historical load data, real - time working condition data and a pre - trained load prediction model, determining the load demand curve of each monitoring sub - region within a preset time period; Inputting the corrected photovoltaic power output, the load demand curve and the multi - dimensional perception data into a pre - constructed dynamic weight collaborative model to determine an output weight triple; the output weight triple includes a photovoltaic power output weight, an energy storage output weight and a grid output weight; When it is determined that the real - time fluctuation of the output weight triple is greater than a preset threshold, based on a preset abnormal level matrix, determining the corresponding weight adjustment parameter and / or generating a cross - sub - region power supply relay path, and storing the corresponding scheduling data in a preset database to update the load prediction model and the weight adjustment parameter; Among them, based on a preset abnormal level matrix, determining the corresponding weight adjustment parameter and / or generating a cross - sub - region power supply relay path specifically includes: Presetting a first preset fluctuation amplitude F1, a second preset fluctuation amplitude F2 and a third preset fluctuation amplitude F3, and F1 < F2 < F3, presetting a first preset fluctuation duration t1 and a second preset fluctuation duration t2, t1 < t2, and constructing the preset abnormal level matrix according to the abnormal levels preset by the user and the combinations of each preset fluctuation amplitude and each preset fluctuation duration; Determining the real - time fluctuation amplitude and the corresponding fluctuation duration corresponding to the output weight triple; Comparing the real - time fluctuation amplitude with each preset fluctuation amplitude in the preset abnormal level matrix, and comparing the fluctuation duration with each preset fluctuation duration in each preset abnormal level matrix to determine the abnormal level according to the comparison results; In the case where the abnormal level is level one, determining the corresponding weight adjustment parameter according to a preset temporary wave suppression correction comparison table to correct each output weight in the output weight triple; the correction includes positive correction and negative correction; In the case where the abnormal level is level two, determining the corresponding weight adjustment parameter according to the preset temporary wave suppression correction comparison table, correcting the local output weight in the output weight triple, and generating the cross - sub - region power supply relay path so that other monitoring sub - regions supply power to the corresponding monitoring sub - region; In the case where the abnormal level is level three, determining the corresponding weight adjustment parameter according to the preset temporary wave suppression correction comparison table, correcting the global output weight in the output weight triple, and generating the cross - sub - region power supply relay path so that other monitoring sub - regions supply power to the corresponding monitoring sub - region.

2. The method for coordinated scheduling of optical and energy storage for smart construction site equipment according to claim 1, characterized in that, Based on the light intensity, ambient temperature, and construction site shading coefficient corresponding to the monitored sub-region, the photovoltaic output is nonlinearly corrected to obtain the corrected photovoltaic output, specifically including: Calculate the light intensity correction parameters based on the light intensity and the light nonlinearity correction coefficient; Calculate the ambient temperature correction parameters based on the ambient temperature and the preset temperature correction coefficient; The corrected photovoltaic output is calculated based on the photovoltaic output, the light intensity correction parameter, the ambient temperature correction parameter, and the construction site shading coefficient.

3. The method for coordinated scheduling of optical and energy storage for smart construction site equipment according to claim 2, characterized in that, Before calculating the corrected photovoltaic output based on the photovoltaic output, the light intensity correction parameter, the ambient temperature correction parameter, and the construction site shading coefficient, the method further includes: Obtain the real-time location coordinates of the tower cranes and the photovoltaic arrays at the smart construction site, and unify them into the global coordinate system of the construction site; Obtain the real-time solar altitude angle and solar azimuth angle from the user terminal, and calculate the unit vector of the solar direction; Based on the real-time location coordinates, the solar direction unit vector, and the preset shadow projection model, the tower shadow area and the crane boom shadow area corresponding to the smart construction site tower crane are determined. Based on the coordinates of the photovoltaic array, the shadow area of ​​the tower body, and the shadow area of ​​the crane arm, the shaded area corresponding to the photovoltaic array is determined, and the site shading coefficient corresponding to the photovoltaic array is determined based on the shaded area and the total area of ​​the photovoltaic array.

4. The method for coordinated scheduling of optical and energy storage for smart construction site equipment according to claim 1, characterized in that, Before determining the load demand curve for each monitoring sub-region within a preset time period based on relevant historical load data, real-time operating condition data, and a pre-trained load prediction model, the method further includes: Determine the region type of the monitoring sub-region corresponding to the load demand curve to be predicted; the region type includes at least hoisting area, mixing area, and auxiliary area; the monitoring sub-region corresponding to the hoisting area includes at least the following smart construction site equipment: tower crane, construction elevator; the monitoring sub-region corresponding to the auxiliary area includes at least the following smart construction site equipment: lighting equipment, monitoring equipment. Based on the area type, a corresponding model channel is determined to call the first prediction sub-model in the load prediction model to perform load prediction for the hoisting area monitoring sub-area, and to call the second prediction sub-model in the load prediction model to perform load prediction for the auxiliary area monitoring sub-area; the computing power required by the first prediction sub-model is at least greater than that of the second prediction sub-model.

5. The method for coordinated scheduling of optical and energy storage for smart construction site equipment according to claim 1, characterized in that, The corrected photovoltaic output, the load demand curve, and the energy storage SOC are input into a pre-constructed dynamic weighted collaborative model to determine the output weight triplet, specifically including: With the goal of satisfying the load demand curve, the dynamic weighted coordination model determines the corresponding photovoltaic output weight based on the corrected photovoltaic output, maximum photovoltaic output, grid stability coefficient, load of each device in the corresponding monitoring sub-region, and priority coefficient of each device; wherein, the device priority coefficient is positively correlated with the energy consumption of the device. The corresponding energy storage output weight is determined by the dynamic weight coordination model based on the energy storage SOC, the load of each device, the priority coefficient of each device, the corrected photovoltaic output, and the maximum energy storage output. The dynamic weighted coordination model determines the grid output weight based on the photovoltaic output weight and the energy storage output weight, and adds the photovoltaic output weight, the energy storage output weight, and the grid output weight to the output weight triplet.

6. A method for coordinated scheduling of optical and energy storage systems for smart construction site equipment according to any one of claims 1-5, characterized in that, Generate a power relay path across sub-regions, specifically including: Construct a smart construction site topology map, wherein the nodes of the construction site topology map are each of the monitoring sub-regions, and the edges are power transmission lines; Based on the energy storage SOC and the maximum preset energy storage output of each node, the energy redundancy corresponding to each node is determined. The transmission loss corresponding to each side is determined based on the resistivity of the transmission line, the distance between each side, and the preset cross-sectional area of ​​the conductor. The abnormal monitoring sub-regions where the real-time fluctuation of the output weight triplet is greater than a preset threshold are taken as target support regions. Based on the load gap value corresponding to the target support region and the energy redundancy corresponding to each monitoring sub-region, one or more candidate support regions are screened in each monitoring sub-region. When there is only one candidate support region, it is determined as the support region. When there are multiple candidate support regions, each path weight is determined based on the energy redundancy and corresponding transmission loss of each candidate support region, so as to select a support region according to the path weight; wherein the sum of the energy redundancy corresponding to multiple support regions is at least greater than the load gap value of the target support region; Based on the edge between the support area and the target support area in the construction site topology map, the cross-sub-region power supply relay path is established.

7. A method for coordinated scheduling of optical and energy storage for smart construction site equipment according to claim 6, characterized in that, Based on the path weights, the supported regions are selected, specifically including: If multiple target support regions are determined to exist, the product of the device priority coefficient corresponding to the target support region and the corresponding path weight is calculated, and the product value is used as the updated path weight to filter out the support regions according to the updated path weight.

8. A photoelectric storage collaborative scheduling system for smart construction site equipment, characterized in that, The system is capable of executing the optical-storage collaborative scheduling method for smart construction site equipment as described in any one of claims 1-7; the system includes: The acquisition module is used to divide the monitoring sub-areas according to the functional zoning of the construction site and the geographical coordinates, and to acquire multi-dimensional sensing data of each monitoring sub-area through a pre-deployed sensor network at a corresponding preset monitoring frequency; wherein, the multi-dimensional sensing data includes at least: photovoltaic output, energy storage SOC, equipment energy consumption, and grid stability coefficient. The correction module is used to perform nonlinear correction on the photovoltaic output based on the light intensity, ambient temperature and construction site shading coefficient corresponding to the monitoring sub-region, so as to obtain the corrected photovoltaic output. The first determining module is used to determine the load demand curve of each monitoring sub-region within a preset time period based on the corresponding historical load data, real-time operating condition data and pre-trained load prediction model. The input determination module is used to input the corrected photovoltaic output, the load demand curve, and the multi-dimensional sensing data into a pre-constructed dynamic weighted coordination model to determine the output weight triplet; the output weight triplet includes photovoltaic output weight, energy storage output weight, and grid output weight. The second determining module is used to determine the corresponding weight adjustment parameters and / or generate a cross-sub-region power supply relay path based on a preset anomaly level matrix when the real-time fluctuation of the output weight triplet is determined to be greater than a preset threshold, and to store the corresponding scheduling data in a preset database to update the load prediction model and the weight adjustment parameters.

9. A photoelectric storage collaborative scheduling device for smart construction site equipment, characterized in that, The device includes: At least one processor; and, A memory communicatively connected to the at least one processor; wherein, The memory stores instructions that can be executed by the at least one processor, which, when executed by the at least one processor, enables the at least one processor to perform a photoelectric storage collaborative scheduling method for smart construction site equipment as described in any one of claims 1-7.

Images