Operation state estimation system, training device, estimation device, state estimator generation method, and estimation method

An operating condition estimation system comprising:

• a learning apparatus that learns a condition estimator which estimates, from information that is acquired while a fluid catalytic cracking apparatus is being operated, an operating condition of the fluid catalytic cracking apparatus, the fluid catalytic cracking apparatus including a reaction apparatus in which a catalyst is used and a regeneration apparatus which regenerates the catalyst;
• and
• an estimation apparatus that estimates the operating condition of the fluid catalytic cracking apparatus by using the condition estimator learned by the learning apparatus based on the information acquired while the fluid catalytic cracking apparatus is being operated, wherein
• the learning apparatus includes a first processor configured to:
• acquire, as learning data, the information acquired when the fluid catalytic cracking apparatus was operated in the past;
• and
• learn the condition estimator through machine learning, using the learning data acquired by the first processor, and
• the estimation apparatus includes a second processor configured to:
• acquire the information acquired while the fluid catalytic cracking apparatus is being operated;
• transform the acquired information into a multi-dimensional data array in which each dimension corresponds to a respective physical parameter of the fluid catalytic cracking apparatus;
• calculate, from the multi-dimensional data array, at least one of a temperature gradient within the regeneration apparatus, a ratio between a carbon monoxide concentration and a carbon dioxide concentration, and a deviation from a reference operating condition;
• generate a feature map by arranging the at least one of the temperature gradient, the ratio, and the deviation into a structured feature vector or matrix representing a physical state of the regeneration apparatus, the feature map including a feature amount having a reduced number of dimensions;
• estimate the operating condition of the fluid catalytic cracking apparatus by inputting the information and the feature map generated by the second processor to the condition estimator;
• output information indicating the operating condition of the fluid catalytic cracking apparatus estimated by the second processor;
• and
• display a chart showing the feature amount calculated by the second processor plotted in a two-dimensional coordinate space or a three-dimensional coordinate space,
• wherein when the information acquired when the fluid catalytic cracking apparatus was operated at a predetermined point of time in the past is input to the condition estimator, the first processor learns the condition estimator so that a predicted value of a different type of information acquired after an elapse of a predetermined period of time since the predetermined point of time is output from the condition estimator,
• wherein the first processor is further configured to adjust a weight of each of a plurality of types of information acquired by the first processor from which information the condition estimator calculates the predicted value, based on a result of fault tree analysis in which an occurrence of afterburn in the regeneration apparatus is defined as an event above, and
• the condition estimator calculates the feature amount having a smaller number of dimensions than the information acquired by the first processor,
• wherein the information includes at least one of:
• a temperature and temperature distribution inside the regeneration apparatus;
• a temperature and quantity of an exhaust gas discharged from the regeneration apparatus;
• a carbon monoxide concentration and a carbon dioxide concentration in the exhaust gas,
• a temperature, a flow volume, and a flow rate of a fluid flowing in a pipe connected to the regeneration apparatus or a riser,
• a position of the valve provided in the pipe,
• a temperature and a quantity of an exhaust gas discharged from the regeneration apparatus,
• a carbon monoxide concentration and a carbon dioxide concentration in the exhaust gas,
• a temperature of the regeneration catalyst supplied from the regeneration apparatus to the riser,
• an amount of coke left;
• a temperature and a quantity of a cracked product supplied from the riser to a stripper;
• and
• a temperature and a quantity of light distillate stripped from the stripper.

An operating condition estimation system comprising:

• a learning apparatus that learns a condition estimator which estimates, from information that is acquired while a fluid catalytic cracking apparatus is being operated, an operating condition of the fluid catalytic cracking apparatus, the fluid catalytic cracking apparatus including a reaction apparatus in which a catalyst is used and a regeneration apparatus which regenerates the catalyst;
• and
• an estimation apparatus that estimates the operating condition of the fluid catalytic cracking apparatus by using the condition estimator learned by the learning apparatus based on the information acquired while the fluid catalytic cracking apparatus is being operated, wherein
• the learning apparatus includes a first processor configured to:
• acquire, as learning data, the information acquired when the fluid catalytic cracking apparatus was operated in the past;
• and
• learn the condition estimator through machine learning, using the learning data acquired by the first processor, and
• the estimation apparatus includes a second processor configured to:
• acquire the information acquired while the fluid catalytic cracking apparatus is being operated;
• transform the acquired information into a multi-dimensional data array in which each dimension corresponds to a respective physical parameter of the fluid catalytic cracking apparatus;
• calculate, from the multi-dimensional data array, at least one of a temperature gradient within the regeneration apparatus, a ratio between a carbon monoxide concentration and a carbon dioxide concentration, and a deviation from a reference operating condition;
• generate a feature map by arranging the at least one of the temperature gradient, the ratio, and the deviation into a structured feature vector or matrix representing a physical state of the regeneration apparatus, the feature map including a feature amount having a reduced number of dimensions;
• estimate the operating condition of the fluid catalytic cracking apparatus by inputting the information and the feature map generated by the second processor to the condition estimator;
• output information indicating the operating condition of the fluid catalytic cracking apparatus estimated by the second processor;
• and
• display a chart showing the feature amount calculated by the second processor plotted in a two-dimensional coordinate space or a three-dimensional coordinate space,
• wherein
• the first processor of the learning apparatus is further configured to generate learning data by adjusting a plurality of types of information acquired by the first processor according to an offset time that depends on the type of information,
• the second processor of the estimation apparatus is further configured to generate input data input to the condition estimator by adjusting a plurality of types of information acquired by the second processor according to the offset time that depends on the type of information, and
• the condition estimator calculates the feature amount having a smaller number of dimensions than the information acquired by the first processor,
• wherein the information includes at least one of:
• a temperature and temperature distribution inside the regeneration apparatus;
• a temperature and quantity of an exhaust gas discharged from the regeneration apparatus;
• a carbon monoxide concentration and a carbon dioxide concentration in the exhaust gas,
• a temperature, a flow volume, and a flow rate of a fluid flowing in a pipe connected to the regeneration apparatus or a riser,
• a position of the valve provided in the pipe,
• a temperature and a quantity of an exhaust gas discharged from the regeneration apparatus,
• a carbon monoxide concentration and a carbon dioxide concentration in the exhaust gas,
• a temperature of the regeneration catalyst supplied from the regeneration apparatus to the riser,
• an amount of coke left;
• a temperature and a quantity of a cracked product supplied from the riser to a stripper;
• and
• a temperature and a quantity of light distillate stripped from the stripper.