Johnson Controls Technology Company

A system includes equipment operable to affect, monitor, or control one or more variable states or conditions in a building, a sensor configured to measure a variable relating to performance of the equipment, and a control system. The control system is configured to perform performance checks for the equipment using a measurement from the sensor and determine individual performance check indicators based on the performance checks. The individual performance check indicators indicate events having different timings and severities. The control system is configured to generate an overall performance index for the equipment using the individual performance check indicators and weights based on the different timings and severities of the events indicated by the individual performance check indicators. The control system is configured to cause an adjustment to the equipment by initiating or executing a preventative maintenance action for the equipment based on the overall performance index.

A model predictive maintenance (MPM) system for building equipment includes one or more processing circuits including one or more processors and memory storing instructions that, when executed by the one or more processors, cause the one or more processors to perform operations. The operations include obtaining one or more performance indicators for the building equipment and determining whether a trigger condition has been satisfied based on the one or more performance indicators. The operations include triggering a model predictive maintenance process to generate a maintenance schedule for the building equipment in response to determining that the trigger condition has been satisfied. The operations include initiating a maintenance activity for the building equipment in accordance with the maintenance schedule.

A method for determining a stability of a building device includes receiving time series data from building equipment regarding a first operational parameter of the building equipment, determining, for one or more time steps of the time series data, a deviation between a setpoint and a recorded value of the first operational parameter, calculating, for the one or more time steps, a trapezoidal area based on an absolute value of the deviation and a length of a time step, calculating a stability index value for the building equipment by summing the trapezoidal areas for a subset of the one or more time steps over a time horizon, and initiating an automated response based on the stability index value.

A heating, ventilation, and air conditioning (HVAC) fault prediction system for a building including a processing circuit including a processor and memory, the memory having instructions stored thereon that, when executed by the processor, cause the processing circuit to receive HVAC data relating to a plurality of HVAC components, the HVAC data indicating performance of the plurality of HVAC components, generate, based on the received HVAC data, a univariate prediction model and a multivariate prediction model, generate, using the received HVAC data, one or more predicted operational parameters for the plurality of HVAC components corresponding to a future time period, and execute at least one of the univariate prediction model or the multivariate prediction model on the one or more predicted operational parameters to predict a HVAC fault associated with at least one of the plurality of HVAC components to occur during the future time period.

A building management system includes connected equipment configured to measure a plurality of monitored variables and a predictive diagnostics system configured to receive the monitored variables from the connected equipment; generate a probability distribution of the plurality of monitored variables; determine a boundary for the probability distribution using a supervised machine learning technique to separate normal conditions from faulty conditions indicated by the plurality of monitored variables; separate the faulty conditions into sub-patterns using an unsupervised machine learning technique to generate a fault prediction model, each sub-pattern corresponding with a fault, and each fault associated with a fault diagnosis; receive a current set of the monitored variables from the connected equipment; determine whether the current set of monitored variables correspond with one of the sub-patterns of the fault prediction model to facilitate predicting whether a corresponding fault will occur; and determining the fault diagnosis associated with the predicted fault.

A building management system includes connected equipment configured to measure a plurality of monitored variables and a predictive diagnostics system configured to receive the monitored variables from the connected equipment; generate a probability distribution of the plurality of monitored variables; determine a boundary for the probability distribution using a supervised machine learning technique to separate normal conditions from faulty conditions indicated by the plurality of monitored variables; separate the faulty conditions into sub-patterns using an unsupervised machine learning technique to generate a fault prediction model, each sub-pattern corresponding with a fault, and each fault associated with a fault diagnosis; receive a current set of the monitored variables from the connected equipment; determine whether the current set of monitored variables correspond with one of the sub-patterns of the fault prediction model to facilitate predicting whether a corresponding fault will occur; and determining the fault diagnosis associated with the predicted fault.

A model predictive maintenance (MPM) system for building equipment includes an operational cost predictor configured to predict a cost of operating the building equipment over a duration of an optimization period, a maintenance cost predictor configured to predict a cost of performing maintenance on the building equipment over the duration of the optimization period, and an objective function optimizer configured to optimize an objective function to predict a total cost associated with the building equipment over the duration of the optimization period. The objective function includes the predicted cost of operating the building equipment and the predicted cost of performing maintenance on the building equipment. The MPM system includes an equipment controller configured to operate the building equipment to affect a variable state or condition in a building in accordance with values of one or more decision variables obtained by optimizing the objective function.

A method performed by a Building Management System (BMS) includes receiving a vibration dataset associated with a machine controlled by the BMS. The vibration dataset includes machine metadata, machine operating conditions, and one or more time waveforms. The method further includes assembling a feature vector comprising one or more features of the vibration dataset for input to a machine learning model. The machine learning model is associated with a component of the machine. The method further includes applying the feature vector to the machine learning model in order to generate a condition score for the component. The method further includes scheduling a maintenance event for the machine or changing a setpoint associated with the machine depending on the condition score.

A method for detecting anomalous behavior includes receiving a plurality of historical events that occurred over a time period, identifying respective timestamps of the plurality of events, and grouping the plurality of events into respective time buckets that extend over a predetermined time interval based on the respective timestamps. The method includes identifying a subset of the time buckets that are each associated with a particular first time unit of the plurality of first time units and a particular second time unit of the second plurality of time units, counting a number of the events in each of the subset, and generating a temporal profile including an occurrence rate calculated based on the counted number of events. The method includes receiving detected event(s) that belong to the subset, calculating a probability of observing the event(s) according to the temporal profile, and determining whether the event(s) correspond to an anomalous behavior by comparing the probability to predefined thresholds.

A controller for performing a peer analysis of a predictive model for a building including processors and non-transitory computer-readable media storing instructions that, when executed by the processors, cause the processors to perform operations. The operations include comparing model parameters of a set of predictive models corresponding to a set of building zones, the model parameters indicating system dynamics of the building zones, to determine whether any of the building zones are exhibiting abnormal system dynamics. The operations include initiating a corrective action in response to determining that at least one of the building zones is exhibiting abnormal system dynamics.