A building system of a building including one or more memory devices having instructions thereon, that, when executed by one or more processors, cause the one or more processors to receive a modification to a graph, the graph comprising a plurality of nodes and a plurality of edges between the plurality of nodes, the plurality of nodes representing entities of the building and the plurality of edges representing relationships between the entities of the building. The instructions cause the one or more processors to generate a change feed event, the change feed event recording the modification to the graph and add the change feed event to a change feed comprising a plurality of change feed events representing modifications to the graph at a plurality of different times.

A building system of a building including one or more memory devices having instructions thereon, that, when executed by one or more processors, cause the one or more processors to receive a request to access a portion of a building graph of the building from a system, the building graph including a plurality of nodes and a plurality of edges, the plurality of nodes representing entities of the building and the plurality of edges representing relationships between the entities of the building. The instructions cause one or more processors to determine whether the system has access to the portion of the building graph based on a policy indicated by the plurality of nodes and edges of the building graph and provide the portion of the building graph to the system in response to a determination that the system has access to the portion of the building graph.

Systems and methods for training a surrogate model for predicting system states for a building management system based on generated data from a system identification model are disclosed herein. The system identification model is used to generate predicted system parameters of a zone of the building based on historic data from operation of the building equipment. The surrogate model is trained based on the predicted system parameters from the system identification model. Predicted future parameters of the variable state of the building are generated using the surrogate model. The surrogate model is re-trained based on new operational data from the building equipment. An updated series of predicted future parameters is generated using the re-trained surrogate model.

A method for automated optimisation of a servo control system controlled by a setpoint, the servo control system including a corrector in a feedback loop, the method exhibiting satisfactory reliability and performance in terms of stability through an iterative procedure, the most effective corrector being determined from among correctors by developing a current value of the delay margin and by individually testing the correctors on the servo control system of the real mechatronic system and by injecting an excitation signal into the loop and by assessing two effective indicators based on at least one effective static margin and one effective dynamic margin, the two effective indicators being an effective static indicator and an effective dynamic indicator, the iterative procedure being stopped on a corrector, which is then the optimal corrector, when the two effective indicators become greater than respective thresholds determined for a current delay margin value.

The present disclosure provides a scheduling method for a power system based on flexible HVDC (high-voltage direct current) and a pumped storage power station, which belongs to a field of power system control technologies. The method is applicable in a power system having a flexible HVDC system and a pumped storage power station. By establishing a scheduling model for the power system, which contains an objective function and multiple constraints, and solving the scheduling model, a capability of the pumped storage power station is used to adjust the unstable output of the renewable energy power generator and stabilize fluctuant of the renewable energy power generation, such that a power incoming into a load center presents a stable ladder pattern and an optimal scheduling scheme can be obtained.

A computer-implemented learning method for optimizing a control policy controlling a system is provided. The method includes receiving states of the system being operated for a specific task, initializing the control policy as a function approximator including neural networks, collecting state transition and reward data using a current control policy, estimating an advantage function and a state visitation frequency based on the current control policy, updating the current control policy using the second-order approximation of the objective function, a second-order approximation of the KL-divergence constraint on the permissible change in the policy using a quasi-newton trust region policy optimization, and determining an optimal control policy, for controlling the system, based on the average reward accumulated using the updated current control policy.

The embodiments described herein include one embodiment that a control method including executing an infeasible search algorithm during a first portion of a predetermined sample period to search for a feasible control trajectory of a plurality of variables of a controlled process, executing a feasible search algorithm during a second portion of the predetermined sample period to determine the feasible control trajectory if the infeasible search algorithm does not determine a feasible control trajectory, and controlling the controlled process by application of the feasible control trajectory.

Machine learning techniques provide predictions of computer system load and cooling requirements, allowing cooling systems to anticipate, prepare for, and ameliorate, those requirements. Techniques provide confirmation or denial that the computer and cooling systems are operating as expected, allowing those systems to determine if any parts are failing. Techniques provide information regarding likely effect due to proposed changes. Techniques provide probabilistic predictions, well in advance, of whether parts will fail or degrade. Techniques provide probabilistic predictions whether parts will fail or degrade, responsive to degrees of redundancy, set points for of cooling, or set points when cooling starts or ends.

A building system of a building including one or more memory devices having instructions thereon, that, when executed by one or more processors, cause the one or more processors to manage a plurality of entitlements for a plurality of subscriptions of one or more buildings with a building entitlement model, receive a first request to perform a first operation for a first subscription and a second request to perform a second operation for a second subscription, and implement the first operation on first computing resources of a first zone based on the building entitlement model in response to a first determination that the first subscription has the first entitlement and implement the second operation on second computing resources of the second zone based on the building entitlement model in response to a second determination that the second subscription has the second entitlement.

A nonlinear disturbance rejection control apparatus and method for electronic throttle control systems are invented to control the electronic throttle system and to achieve a continuous finite-time disturbance rejection control goal. A control sub-apparatus and method are proposed with an observing sub-apparatus and method for controlling the opening angle of an electronic throttle valve. A mathematical model of the electronic throttle system is analyzed and a control-oriented model is presented with the formation of a lumped disturbance. With combination of the continuous terminal sliding mode control method and the output feedback control method, based on the finite-time high-order sliding mode observer, the preferred control performance is guaranteed, where both the dynamic and static performance of the system is effectively improved.